Enabling Iron‐Based Highly Effective Electrochemical Water‐Splitting and Selective Oxygenation of Organic Substrates through In Situ Surface Modification of Intermetallic Iron Stannide Precatalyst [original]

www.advenergymat.de

Full PaPer

Enabling Iron-Based Highly Effective Electrochemical

Water-Splitting and Selective Oxygenation of Organic

Substrates through In Situ Surface Modification of

Intermetallic Iron Stannide Precatalyst

Biswarup Chakraborty, Rodrigo Beltrán-Suito, J. Niklas Hausmann, Somenath Garai,

Matthias Driess,* and Prashanth W. Menezes*

DOI: 10.1002/aenm.202001377

1. Introduction

Electrochemical water-splitting through

the hydrogen evolution reaction (HER)

and oxygen evolution reaction (OER) has

been regarded as one of the most promi-

sing ways to generate hydrogen (H2), a

renewable fuel.[1] However, the efficiency

of water splitting is largely limited by

the thermodynamically up-hill OER that

requires multiple proton-coupled elec-

tron transfer steps with complex reaction

kinetics.[2] Consequently, large overpo-

tentials are required to achieve the OER

process making it the bottleneck of water

splitting.[3] In order to reduce the energy

barrier of OER, efficient and durable elec-

trocatalysts (anodes) that can display low

overpotentials at higher current densities

with fast reaction kinetics are desired.[4]

Although the electrochemical water split-

ting to produce oxygen (O2) and H2 with

noble metal-based catalysts (RuO2, IrO2

for the OER and Pt for HER) currently

remains successful, the design of cost-

effective, active and sustainable materials utilizing natural

resources of nonprecious metals begins to be a central thrust

in the advancement of OER and HER electrocatalysts.[5] Recent

success with Earth-abundant metal-based electro(pre)catalysts

for water splitting mostly rely on transition metal-oxides,[6]

phosphates,[7] chalcogenides,[8] pnictides,[9] borides,[10] phos-

phites,[11] and borophosphates.[12]

Within the transition-metals, iron is the naturally most abun-

dant in Earth’s-crust.[13] Besides its abundance, accessible mul-

tiple redox states and rich coordination chemistry has made iron

a biologically significant metal ion and it is present as co-factor

in many metalloenzymes.[14] Taking this into account, several

iron-based materials have been designed as photo[15] and elec-

trocatalysts for OER, especially, for alkaline water splitting,[16]

HER[17] and nitrogen reduction.[18] However, poor electrical

conductivity,[16a] the tendency of iron to leach into the solution

as a thermodynamically stable soluble (FeO4)2− species (as pre-

dicted by Pourbaix diagram),[19] and a nonoptimal oxophilicity of

iron in comparison to other transition metals limits its efficiency

A strategy to overcome the unsatisfying catalytic performance and the dura-

bility of monometallic iron-based materials for the electrochemical oxygen

evolution reaction (OER) is provided by heterobimetallic iron–metal systems.

Monometallic Fe catalysts show limited performance mostly due to poor

conductivity and stability. Here, by taking advantage of the structurally ordered

and highly conducting FeSn2 nanostructure, for the first time, an intermetallic

iron material is employed as an efficient anode for the alkaline OER, overall

water-splitting, and also for selective oxygenation of organic substrates. The

electrophoretically deposited FeSn2 on nickel foam (NF) and fluorine-doped

tin oxide (FTO) electrodes displays remarkable OER activity and durability

with substantially low overpotentials of 197 and 273mV at 10mA cm−2,

respectively, which outperform most of the benchmarking NiFe-based cata-

lysts. The resulting superior activity is attributed to the in situ generation of

α-FeO(OH)@FeSn2 where α-FeO(OH) acts as the active site while FeSn2

remains the conductive core. When the FeSn2 anode is coupled with a Pt

cathode for overall alkaline water-splitting, a reduced cell potential (1.53V) is

attained outperforming that of noble metal-based catalysts. FeSn2 is further

applied as an anode to produce value-added products through selective oxy-

genation reactions of organic substrates.

Dr. B. Chakraborty, R. Beltrán-Suito, J. N. Hausmann,

Prof. M. Driess, Dr. P. W. Menezes

Department of Chemistry: Metalorganics and Inorganic Materials

Technische Universität Berlin

Straße des 17 Juni 135, Sekr. C2, Berlin 10623, Germany

E-mail: [email protected]; prashanth.menezes@mailbox.

tu-berlin.de

Dr. S. Garai

Department of Chemistry

National Institute of Technology

Tiruchirappalli, Tamil Nadu 620015, India

The ORCID identification number(s) for the author(s) of this article

can be found under https://doi.org/10.1002/aenm.202001377.

KGaA, Weinheim. This is an open access article under the terms of the

Creative Commons Attribution-NonCommercial-NoDerivs License,

which permits use and distribution in any medium, provided the original

work is properly cited, the use is non-commercial and no modifications

or adaptations are made.

Adv. Energy Mater. 2020, 10, 2001377

www.advenergymat.de

www.advancedsciencenews.com

in alkaline OER catalysis.[20] Such shortcomings can be over-

come by the incorporation of iron into nickel[21] or cobalt oxyhy-

droxides[22] and currently, the mixed nickel–iron oxyhydroxides

(NixFe1−xOOH) are considered to be the most promising noble

metal-free OER catalysts.[23] Although the nature of the active

site and the role of iron in the promotion of catalytic efficiency

is still debatable,[24] in-depth spectro- electrochemical, in situ and

operando spectroscopic,[25] as well as theoretical studies, predict

a high-valent iron as the active species.[26] Hence, unconven-

tional catalysts based on iron that are strikingly active, highly

conductive, and can sustain in the operating conditions by dis-

favouring iron dissolution are urgently needed to fulfil the void

required for the next generation OER.[27]

Similar to OER, electrooxidation of organic substrates to

valuable chemicals has been regarded as a potential alternative

to avoid hazardous chemical-oxidants and harsh reaction condi-

tions. In this context, electrosynthesis of 2,5-furandicarboxylic

acid (FDCA)[28] is a value-added precursor for polymers such as

polyethylene terephthalate and poly(ethylene 2,5-furandicarbo-

xylate) and a convenient replacement of terephthalic acid. The

electrooxidation of FDCA via selective oxidation of 5-hydroxy-

methylfurfural (HMF)[29] is limited to noble metal-based elec-

trodes.[30] Although recent efforts to replace the noble-metal

catalysts with transition metal-based materials have been

made, the selectivity and efficiency of electrosynthesis of these

biomass-derived organic products remain a great challenge.[31]

On the other hand, electrooxidation of ethanol to acetic acid,

a high boiling solvent commonly used for organic synthesis,

could also be efficiently performed only with noble-metal elec-

trodes.[32] Therefore, it is of significant interest to develop low-

cost, non-noble metal-based, and high-performance catalysts

that can unify both the kinetically sluggish OER and oxygena-

tion reactions at the anode as well as continuous production of

H2 at the cathode by making the complete system energetically

efficient.

Over the years, structurally ordered intermetallic compounds

have gained enormous attention because of their unique crys-

tallographic, chemical, physical and electronic properties,

such as magnetism and superconductivity.[33] As the arrange-

ments of atoms in an intermetallic compound can be tailored

to achieve superior electronic and adsorption properties, the

interest has even grown in the last couple of years to explore

intermetallics as active catalysts for application in electro-

catalytic reactions.[34] Within intermetallic compounds, metal

stannides (MnSn2, FeSn2, and CoSn2) have drawn special atten-

tion, as they possess a distinct connection pattern between the

transition metal and tin atoms through covalent interactions

leading to high electrical conductivity (metal-like character). In

this regard, they have already been used successfully as anode

materials for Li-ion batteries.[35] Notably, intermetallics based

on noble and non-noble metals have recently been utilized

successfully for the reaction of OER, HER, and oxygen reduc-

tion reaction (ORR).[36] Along this line, lately, a CoSn2 nano-

structure acting as a bifunctional electro(pre)catalyst for OER,

HER and overall water-splitting in alkaline media has been

reported resulting in substantially low overpotentials and high

long-term stabilities.[37] Furthermore, an improved electrocata-

lytic OER performance in both alkaline and neutral electrolyte

has also been achieved after selenization of Cu3Sn@Cu.[38]

Motivated by these promising results, we aimed to investigate

iron stannides as a promising class of anode materials for the

kinetically demanding reactions of both OER and selective

oxygenation.

Here, we report the synthesis of structurally ordered inter-

metallic iron stannide (FeSn2) nanocrystals and apply them as

highly active OER electro(pre) catalysts. The electrodeposited

FeSn2 on both fluorine-doped tin oxide (FTO) and nickel foam

(NF) delivers superior OER performance in alkaline solutions.

A detailed postcatalytic study infers that under electrochemical

conditions, FeSn2 endures surface restructuring and forms in

situ α-FeO(OH)@FeSn2 as the active catalyst. The superior per-

formance of α-FeO(OH)@FeSn2 over the similarly prepared

Fe, reference Fe(OH)3, α-FeO(OH), benchmark NixFe1−xO4 as

well as state-of-the-art RuO2 and IrO2 catalysts highlights the

essential role of the α-FeO(OH) surface as the active species

and FeSn2 as the conductive site to facilitate enhanced O2 pro-

duction. Moreover, a two-electrode cell constructed using Pt as

the cathode and FeSn2 as the anode (Pt(−)//(+)FeSn2/NF) also

displayed a low cell potential for alkaline water electrolysis.

Encouraged by this, we further utilize the FeSn2 anode for the

challenging electrooxidation reactions of i) ethanol to acetic

acid, ii) acetaldehyde to acetic acid, and iii) 5-HMF to 2,5-FDCA,

yielding high efficiencies.

2. Results and Discussion

Structurally ordered intermetallic FeSn2 nanocrystals were

synthesized by a one-pot reduction of iron and tin precursor

using sodium borohydride (NaBH4) in ethylene glycol (EG) at

170°C (see the Experimental Section in the Supporting Infor-

mation for details). At elevated temperature, the slow growth

of intermetallic nanocrystals transpires by the interspersing

of Fe into the interlayers of metallic Sn through a Kirkendall

process (Scheme1).[39] Interestingly, this diffusion-controlled

growth mechanism was implied to be the operative mechanism

to isolate highly crystalline intermetallic stannide nanostruc-

tures from tetragonal β-Sn, where the shape of the nanostruc-

ture was controlled by the diffusion rate of the species.[40] The

solid-state structure of the as-prepared FeSn2 was further evalu-

ated by powder X-ray diffraction (PXRD) where the sharp reflec-

tions matched (Figure1a) clearly with FeSn2 (JCPDS 25-415)

(Figure S1, Supporting Information). FeSn2 crystallizes in the

CuAl2 structure type and belongs to the tetragonal, space group

I4/mcm (space group No. 140) with the lattice parameters: a=

6.539(2) Å and c= 5.325(2) Å. The crystal structure consists of

Fe atoms forming chains parallel to the [001].[41] Each Fe atom

is at the center of a square antiprism formed by adjacent Sn

atoms while each Sn atom has four nearest Fe neighbors, which

are part of two neighboring chains (Figure S2, Supporting

Information).[42] Such building units are further connected

by FeSnFe bonds to produce a 3D network (Scheme1 and

Figure1a, inset). The structure has also been often described

based on the homoatomic Sn−Sn interactions as a network

of tetrahedral stars or layered nets as well as interpenetrating

graphite-like nets.[52]

Transmission electron microscopic (TEM) images depicted

the nanoscale features of the FeSn2 particles (Figure S3,

Adv. Energy Mater. 2020, 10, 2001377

www.advenergymat.dewww.advancedsciencenews.com

Supporting Information), while high-resolution TEM analyses

verified the crystallinity of the nanoparticles with an interplanar

d-spacing around 0.23nm corresponding to the (112) plane of

FeSn2 (Figure1b,c). High crystallinity and phase-purity of the

as-prepared FeSn2 was also evidenced from the well-defined

diffraction rings resolved in the selected area electron diffrac-

tion (SAED) pattern (Figure1c). Scanning electron microscopic

(SEM) images of the FeSn2 (Figure1d; Figure S4, Supporting

Information) revealed aggregation of the monodispersed

nanoparticles. Homogeneous distribution of Fe and Sn in the

particle was confirmed by energy-dispersive X-ray spectroscopy

(EDX) elemental mapping (Figure1d–f). A trace (ca. 1.91%) of

oxygen was also observed during EDX (Figure S5, Supporting

Information) mapping and EDX analysis (Figures S6 and S7,

Supporting Information) which plausibly occurred due to slow

oxidation of FeSn2 in air. The chemical composition of the

FeSn2 was analyzed by inductively coupled plasma atomic emis-

sion spectroscopy (ICP-AES) that confirmed the one to two

Scheme 1. Schematic representation of the formation of structurally ordered intermetallic FeSn2 and close packing of the individual atomic layers of

Fe and Sn (tetragonal, I4/mcm). The ordered structure of FeSn2 is built up via the intercalation of metallic Fe layers (red spheres) and β-Sn layers (grey

spheres) at elevated temperature via the Kirkendall process.[43] Under alkaline electrochemical reaction conditions, a substantial restructuring occurs at

the surface due to the slow depletion of Sn from the FeSn2 nanostructure and subsequent in situ formation of highly active crystalline α-FeIIIO(OH)@

FeSn2 to catalyze both OER and multiple organic oxidations.

Figure 1. Microscopic analyses of the as-prepared FeSn2. a) PXRD pattern of the FeSn2 with the assignment of diffraction planes according to the

reported structure (inset) of FeSn2 (JCPDS 25-415; Figure S1, Supporting Information). b) The TEM image of FeSn2 particles and the respective high-

resolution image with atomic fringes corresponding to (112) crystal lattice plane with an interplanar spacing of 0.23nm (b, inset). c) SAED pattern of

the nanoparticles displaying diffraction rings to the (200), (002), (211), and (202) corresponding to FeSn2. d) SEM image and the elemental mapping

of the FeSn2 displaying the homogenous distribution of Fe. e) and Sn f) shown in yellow and green (oxygen mapping is presented in Figure S5 in the

Supporting Information), respectively.

Adv. Energy Mater. 2020, 10, 2001377

www.advenergymat.de

www.advancedsciencenews.com

ratio of Fe to Sn (Table S1, Supporting Information). Further,

X-ray photoelectron spectroscopy (XPS) was used to investigate

the chemical states and the composition of the elements

(detailed discussion is given in Figures S8 and S9, Supporting

Information). Fe 2p3/2 and Fe 2p1/2 displayed two peaks at the

binding energies of 707.14 and 720.23 eV, corresponding to

Fe0 while peaks of FeIII at 710.65eV (Fe 2p3/2) and 724.36eV

(Fe 2p5/2) were also observed resulting from surface passivation

of the FeSn2 (Figure S8, Supporting Information).[44] Similarly,

Sn 3d5/2 and 3d3/2 exhibited two sharp peaks at the binding ener-

gies of 484.8 and 493.2eV, which could directly be attributed

to the Sn0 whereas the peaks at higher binding energies were

ascribed to the presence of surface oxidized SnII (Figure S9,

Supporting Information).[37,44c,45] To establish a direct electro-

chemical and structure–activity relationship, the metallic Fe

phase was synthesized separately (without the Sn precursor)

in a similar way as that of FeSn2 and characterized through

PXRD, TEM, high-resolution-TEM, SAED, SEM, EDX, and

XPS (Figures S10–S14, Supporting Information).

To evaluate the OER catalytic activity, FeSn2 was electro-

phoretically deposited (EPD) on the FTO electrode surface

to perform electrochemical measurements in a 1 m KOH

aqueous electrolyte in a three-electrode setup. Prior to the

electrochemical measurements, the chemical integrity of the

FeSn2 after depositions was confirmed by various methods

(Figures S15–20, Supporting Information). In order to have

a fair comparison, the metallic Fe, Fe(OH)3 and α-FeO(OH)

were also prepared and electrodeposited with the same mass

loading (for synthetic details see the Experimental Section,

Figures S10–S14 and Figures S21 and S22, Supporting Infor-

mation). Figure2a shows a typical linear sweep voltammetry

(LSV) curve between 1.1 and 1.9V versus reversible hydrogen

electrode (RHE) at a sweep rate of 1mV s−1. The FeSn2/FTO

electrode delivered a much higher OER current density relative

to all other electrodes, indicating the essential role of active Fe

sites and conductive Sn sites in the enhancement of the OER

catalytic activity. Strikingly, an overpotential of only 273 (±6)

mV at a current density of 10mA cm−2 was obtained for FeSn2/

FTO electrode, while at the same current density, the Fe/FTO,

Fe(OH)3/FTO and α-FeO(OH)/FTO displayed overpotentials

of 480 (±5), 576 (±5), and 620 (±6) mV, respectively. To verify

the effect of mass loading and film thickness on the OER

activity of the FeSn2, the electrodeposition time was varied

(from 4 to 240 s) which yielded different loadings of the cata-

lyst on the FTO surface (Table S2, Supporting Information).

Additionally, polarization curves were recorded with different

catalyst loaded films on FTO which also indicated a best OER

activity (Figure S23, Supporting Information) for a loading of

0.8 mg FeSn2 with a thickness of 7.56 (±0.39) µm (Figure S24,

Supporting Information) with a recorded overpotential 273 (±6)

mV at a current density of 10mA cm−2 (Figure3b). It showed

an optimum deposition time of 120 s, after which an increase

in the mass loading was observed without significant change

in the activity (Figure 2b; Table S2, Supporting Information).

Figure 2. Electrocatalytic OER activity with FeSn2/FTO in comparison to Fe/FTO, Fe(OH)3/FTO, and α-FeO(OH)/FTO. a) Polarization curves of FeSn2/

FTO compared to Fe-based reference materials recorded in 1 m KOH solution with a sweep rate of 1mV s−1. b) The overpotentials of FeSn2 on 1 cm2

FTO surface with respect to deposition time of 4 to 240 s (varying mass loadings; see Table S2 in the Supporting Information). c) Nyquist plots (an

equivalent circuit is shown in the inset) at an applied potential of 1.55V versus RHE. d) chronoamperometric OER studies of FeSn2/FTO and Fe/FTO

at a constant potential of 1.53 and 1.70V versus RHE, respectively.

Adv. Energy Mater. 2020, 10, 2001377

www.advenergymat.dewww.advancedsciencenews.com

The turnover frequency (TOF) as a function of the loading

was determined and is also presented in Table S2 in the Sup-

porting Information. The TOF increases with decreased

loading revealing that mass/electron transport phenomena are

relevant and cause a deviation of a linear relationship between

the amount of active sites and the O2 production rate. Achieved

low overpotential with a substantial amount of mass loading

indicates the activity is not arisen due to the overloading of

the catalysts.[16d,46] A comparative mass normalized activity and

TOF calculated per Fe also showed the remarkable OER activity

of FeSn2 (Figure S25, Supporting Information).[23a,47] Further-

more, the catalytic OER performance of FeSn2/FTO was much

higher than that of benchmark RuO2/FTO and IrO2/FTO cata-

lysts as well as other Fe-based materials reported (Figure S26

and Table S3, Supporting Information).[16a–c,e,f,j,19a,48] Notably,

the LSV curve without iR correction gave an overpotential of

316 mV at 10 mA cm−2 which is also better than previously

reported iron-based catalyst (Figure S27 and Table S3, Sup-

porting Information).[49]

The catalytic OER kinetics of the presented catalysts is evalu-

ated by the Tafel plots. A Tafel slope of 33 (±2) mV dec−1 was

determined for FeSn2/FTO, which is comparable to those of

the state-of-the-art benchmark Fe-based catalysts (Figure 2b;

Figure S28, Supporting Information), suggesting facile electron

transfer with intermetallic FeSn2.[49] A semicircular Nyquist plot

obtained from electrochemical impedance spectroscopy (EIS)

measurements with FeSn2 indicated a favorable charge transfer

between electrolyte and electrode interface with a low charge

transfer resistance (RCT) (Figure2c; Table S4, Supporting Infor-

mation).[50] As expected, the RCT values of the presented catalysts

are in accordance with the corresponding electrocatalytic OER

activity. Apart from low overpotentials, which determine the effi-

ciency of a catalyst, stability is another crucial factor to evaluate

its practical implication. Under a chronoamperometric OER con-

dition (OER CA), the FeSn2/FTO catalyst remained stable with

a current density above 15mA cm−2 by an applied potential of

1.53V versus RHE (at an overpotential of 300mV) (Figure2d).

However, a significant drop in the current was observed (up to

61%), when metallic Fe/FTO film was applied under similar

OER CA conditions at 1.7V versus RHE (the necessary potential

to reach 10mA cm−2) making it an unsuitable anode material

most likely due to the low electrical conductivity less abundant

surface active sites compared to FeSn2 (Figure2d).

Dynamic restructuring of the (pre)catalyst under alka-

line electrocatalytic (OER CA) conditions has recently been

studied systematically to reveal the bulk and surface-active

species of the materials and to understand their reaction

pathways.[12,23e,51] Therefore, to further investigate the reason

for the better electrochemical characteristics of FeSn2 over

other catalysts, we examined the FeSn2/FTO films after OER

CA conditions. Foremost, ICP-AES revealed a substantial loss

of Sn (≈30%) suggesting that FeSn2 is indeed an electro(pre)

catalyst that undergoes a structural transformation during the

electrocatalytic OER condition. High-resolution TEM images

showed a crystalline core of FeSn2 surrounded by the crystal-

line α-FeO(OH) layer. The two phases could be differentiated

by the d(112) planes of FeSn2 (JCPDS 25-415) and d(111) planes

of α-FeO(OH) (JCPDS 29-713) near the edge of on an individual

particle (Figure 3a; Figure S29, Supporting Information).[16h]

Furthermore, the SAED pattern depicts the well-defined rings

Figure 3. Post OER CA (24 h) analyses of the FeSn2/FTO. a) The high-resolution TEM image of FeSn2 showing crystalline fringes of d(111) planes of

α-FeO(OH) near the edge of the particle and d(112) planes of FeSn2 at the core. b) SAED pattern obtained from the area shown in a), exhibiting well-

defined rings for FeSn2 and α-FeO(OH). c–f) SEM image (Figure S30, Supporting Information) and EDX elemental mapping of the FeSn2/FTO film

clearly evidencing the presence of Fe (yellow), O (red), and Sn (green) in the material after catalysis. g) Raman spectra of the FeSn2/FTO before (black)

and after OER CA (red). h) PXRD pattern of the FeSn2/FTO before (black) and after OER CA (red); the reflections marked by red asterisk (*) can be

indexed to (110), (120), (130), and (140) planes of a-FeO(OH) phase (JCPDS 29-713).

Adv. Energy Mater. 2020, 10, 2001377

www.advenergymat.de

www.advancedsciencenews.com

of both FeSn2 and α-FeO(OH) (Figure 3b). The SEM image

of FeSn2 (Figure3c) after OER CA displayed noticeable mor-

phological changes during the electrochemical performance

(Figure S30, Supporting Information). The elemental mapping

and EDX (Figure3d–f; Figure S31, Supporting Information) of

the film confirmed the incorporation of an enormous amount

of oxygen into the surface of FeSn2 which further validates the

formation of α-FeO(OH). Raman spectroscopy, a very sensi-

tive probe to surface structure, was also conducted to unveil

the surface structure of FeSn2 before and after OER electroca-

talysis (Figure 3g). The spectrum shows Raman vibrations at

248 (w), 304 (s), 389 (s), 484 (w) and 560 (w) cm−1 which are

in good agreement with the reported vibrations of pure phase

goethite, α-FeO(OH)[52] whereas the Raman spectrum of FeSn2

does not possess any peaks (Figure3g; Figure S32, Supporting

Information). Further analysis of the film by PXRD after OER

CA resulted in sharp reflections for α-FeO(OH) along with the

diffractions for FeSn2 (Figure3h). Additional evidence for the

formation of α-FeO(OH)@FeSn2 under electrocatalytic

condition, via surface restructuring, was provided by XPS. The

Fe 2p XPS spectrum reveals that FeIII is the dominant spe-

cies (Figure S33a, Supporting Information)[44] while a weak

and broad peak corresponding to Fe0 was still retained at the

FeSn2 surface after OER. The Sn 3p and Sn 3d XPS spectra also

exhibited oxidized Sn species on the surface of the film owing

to the trapping of oxidized species from the electrolyte and are

consistent with literature reports (Figure S33a,b, Supporting

Information).[37,44c,45] Likewise, the peaks in the O 1s XPS spec-

trum also support the presence of –OH− and O−II species on

the surface after OER (Figure S33c, Supporting Information).

Overall, the above results evidence the formation of a reactive

α-FeO(OH) overlayer at the surface of the highly conducting

FeSn2 core to achieve efficient OER.

To get additional insight into the reasons for the improved

electrocatalytic performance of FeSn2 among the other Fe-based

catalysts, we performed ex situ four-point probe resistivity

measurements of the thin films. The as-deposited FeSn2/FTO

thin film had a resistivity of 29×101Ω cm. The resistivity of

the surface oxidized, as deposited Fe/FTO had a three to four

magnitudes higher resistivity whereas much higher values

were found for Fe(OH)3/FTO and α-FeO(OH)/FTO (Table S5,

Supporting Information). Surprisingly, α-FeO(OH)@FeSn2/

FTO film was still dramatically more conducting than the oxi-

dized iron compounds (Table S5, Supporting Information). It

should be noted that the ex situ conductivity measurements are

not necessarily able to predict the electron transport abilities

of materials under OER conditions.[53] Nevertheless, conduc-

tivity limitations of FeO(OH) are reported in the literature even

at OER conditions,[19a] and the herein performed conductivity

measurement proves that the remaining FeSn2 can dramati-

cally reduce these limitations even without an applied poten-

tial. Furthermore, the electrocatalytic activity of FeO(OH) is a

strong function of the underlying substrate and the synergistic

interaction of Au/AuOx with FeO(OH) can strongly increase its

OER performance.[19a,54] For α-FeO(OH)@FeSn2, a similar syn-

ergistic effect at the interface between the metallic phase con-

taining the heavy nucleus Sn and α-FeO(OH) might be present.

However, detailed theoretical investigations are needed to prove

this hypothesis. Additionally, chemisorb water is suspected

to block the active iron sites from reactive hydroxide in

FeO(OH).[52c] Such chemisorbed water could also be detected in

high vacuum XPS measurements. In the case of α-FeO(OH)@

FeSn2, no chemisorbed water is present in the O1s XPS indi-

cating good active site availability.

Most importantly, no decrease in the OER activity or leaching

of iron from FeSn2 under CA could be observed in our case. One

reason for this is the low operation potential that is required

to achieve the OER. For amorphous FeO(OH), iron leaches

relatively fast at an overpotential of 450mV, but significantly

slower already at an overpotential of 350mV.[19a] Furthermore,

it was recently shown that crystalline α-FeO(OH), in contrast

to amorphous phases, is stable under OER conditions and

can resist solvation at overpotentials as high as 800mV.[52c] In

the same report, a comparable activity of crystalline α-FeO(OH)

and the amorphous γ-FeO(OH) phase was revealed. This shows

that crystalline phase α-FeO(OH) can combine a high activity

with a good stability, in contrast to previous reports where

amorphous phases are usually considered to be more active.[55]

After the successful demonstration of FeSn2/FTO as a high-

performance electrocatalyst, we deposited the same catalysts on

3D-interconnected porous and conducting NF, without altering

its chemical integrity (Figures S34–S36, Supporting Informa-

tion). The electrocatalytic study was then investigated with

respect to the OER in 1 m KOH aqueous electrolyte. Figure4a

exhibits representative LSV curves of various Fe-based catalysts

on NF at a scan rate of 1mV s−1, along with bare NF as a refer-

ence. Similar to FTO deposited substrates, FeSn2/NF displayed

an exceptionally small overpotential value of 197±4mV at a

current density of 10mA cm−2 and reached an extremely high

catalytic current density of 500mA cm−2 at an electrode poten-

tial of 1.56V (vs RHE) whereas the catalysts Fe/NF, Fe(OH)3/

NF, α-FeO(OH)/NF and bare NF displayed poor catalytic OER

activity. Moreover, the overpotential recorded for the FeSn2/NF

electrocatalyst not only supersedes the state-of-the-art noble

metal catalysts, RuO2/NF and IrO2/NF (Figure S37, Supporting

Information), but this value is significantly lower (33mV) than

that reported analogous intermetallic phase CoSn2 (230 mV

@10mA cm−2) (Figure4b; Table S3, Supporting Information)

and other Fe, Fe–Ni- and Fe–Co-based electrocatalysts (Table S3,

Supporting Information).[56] The recorded overpotential of

FeSn2 is even ca. 130 mV less than the 2D nanoplate FeSe2 cata-

lysts.[57] Notably, an LSV curve without iR correction yielded an

overpotential of 209mV at 10mA cm−2 (Figure S38, Supporting

Information). The Tafel slope calculated for FeSn2 was

31 ±1mV dec−1, which is comparable to the other Fe-based cata-

lysts (Figure4c; Figure S39, Supporting Information).[49] FeSn2/

NF showed a significantly lower RCT (Figure 4d; Figure S40,

Supporting Information) than other electrodes, which high-

lights rapid charge transfer kinetics at the solution–electrode

interface of FeSn2 (Table S6, Supporting Information).[50a]

A detailed analysis was performed to ensure that the higher

electrochemical surface area (ECSA) is most plausibly one of

the controlling factors to obtain the better electrochemical per-

formance of the FeSn2/NF. The calculated double-layer capaci-

tance (Cdl) and the estimated ECSA[58] of FeSn2, and other iron-

based catalyst is summarized in Figure S41 in the Supporting

Information, which confirmed a Cdl of 0.85 mF cm−2 and an

ECSA of 0.49 cm2 forFeSn2 electro(pre)catalyst.[9b,16f,37,51b,59] The

Adv. Energy Mater. 2020, 10, 2001377

www.advenergymat.dewww.advancedsciencenews.com

ECSA normalized electrocatalytic OER activity of the materials

shows that the highest activity is achieved with FeSn2 (Figure

S42, Supporting Information). The Brunauer–Emmet–Teller

(BET) surface areas of the materials have been also determined

(Table S7, Supporting Information), and consequently, the nor-

malization of the activities by these values confirms the best

activity of FeSn2 among the other iron-based catalysts tested

herein (Figure S43, Supporting Information).

The FeSn2/NF sustained for 24 h under an electrochemical

OER CA condition with a subtle change in current density,

very similar to that observed for the OER CA with FeSn2/FTO

(Figure 4f). The LSV curves recorded after OER CA (24 h)

showed almost unchanged overpotentials at 10 mA cm−2

(Figure4f, inset). Post catalytic analyses of FeSn2/NF after OER

CA also indicated the formation of α-FeO(OH) layer covering

the surface of FeSn2 (Figures S44 and S45, Supporting Informa-

tion) and is consistent with surface rearrangements observed

on FeSn2/FTO (Figure3). ICP-AES analysis suggests the spon-

taneous dissolution of Sn from the catalyst’s surface. The loss

of Sn continues with the progress of OER CA and then prob-

ably ends after a certain ratio, as the electrolyte cannot penetrate

to the FeSn2 core anymore forming a stable FeO(OH)@FeSn2.

After 60 h of continuous electrolysis (OER CA) at 220mV over-

potential, ICP-AES analysis of the electrolyte shows that approx-

imately 40% of the Sn leaches out to the electrolyte solution. A

comparatively stable current density was observed throughout

the OER CA measurement (Figure S46, Supporting Informa-

tion). The iron content in the electrolyte was not significantly

above the detection limit of the ICP-AES analysis. The Cdl and

ECSA measured post-OER catalysis showed a slight enhance-

ment that could be attributed to the initial surface conversion

of the FeSn2 (Figure S47, Supporting Information) suggesting

a disordered α-FeO(OH) surface. Moreover, such rigorous

surface modulation/restructuring to form active surface sta-

bilized Fe(OH)2/FeO(OH) electrodes under OER conditions

have recently been realized for Fe-based chalcogenides and

pnictides.[9b,16c,e,f]

A plausible pathway of α-FeO(OH) formation on the FeSn2

surface can be explained via oxidative leaching of Sn, and sub-

sequent oxidation of iron site.[60] As suggested by the Pourbaix

diagrams, during the alkaline OER, a reactive FeII(OH)2 may

form via oxidation of Feδ+ to FeII and that finally transforms

to α-Fe(O)OH (Equation (1)) under applied potentials (note

that at higher anodic applied potentials, the transformation of

FeO(OH) into soluble (FeO4)2− is inevitable).[61]

Sn 15OH Fe OOH2Sn OH 11

III IV

() ()

+→ +



++

−

− (1)

It could be anticipated that a higher degree of structural

ordering of FeSn2 core allows a regular arrangement of Fe(O)

OH which essentially leads to a crystalline nature of the Fe(O)

OH. The conversion ratio of FeSn2 to α-FeO(OH) is strictly con-

trolled by the dissolution rate of Sn and duration of OER. This

was further confirmed by ratios obtained from the EDX and

ICP-AES analysis after the long-term stability tests of 24 and

60 h.

Figure 4. Electrocatalytic OER with FeSn2/NF in comparison to Fe/NF, Fe(OH)3/NF and α-FeO(OH)/NF. a) Polarization curves of the OER recorded

in 1 m KOH solution with a sweep rate of 1mV s−1 b) Comparison of the overpotential recorded at 10mA cm−2 presented materials with respect to

the RuO2/NF and IrO2/NF and recently reported CoSn2/NF, Co, and Sn. c) Tafel slope of FeSn2/NF calculated by the steady-state method. d) Nyquist

plots of Fe based materials an applied potential of 1.55V versus RHE. e) Cdl values of the materials calculated from the slopes of the linear fitting of

Δj (mA cm−2) versus scan rate (mV s−1) (Figure S41, Supporting Information). f) Long-term stability experiment of FeSn2/NF at 1.45V versus RHE and

the inset shows LSV curves before (black) and after OER (red).

Adv. Energy Mater. 2020, 10, 2001377

www.advenergymat.de

www.advancedsciencenews.com

To prove that the OER activity of FeSn2 is superior to the

state-of-the-art NiFe-based catalysts,[24b,c,25a,26b,62] we addition-

ally synthesized a highly active FeNi2O4 phase (see synthesis,

Figure S48, Supporting Information). The FeNi2O4 was elec-

trodeposited on both FTO and NF and measured in alkaline

electrolyte, similar to that of FeSn2. The overpotentials of

FeNi2O4/FTO and FeNi2O4/NF attained at a current density of

10 mA cm−2 were 334 and 230 mV, respectively (Figure S49,

Supporting Information), which are surprisingly higher than

the presented FeSn2 catalysts. This indeed demonstrates the

essential role of Fe as an active site and Sn as a conductive ele-

ment in enhancing the overall catalytic efficiency, without the

introduction of a second transition metal as reported for NiFe-

based materials.[24b,c,25a,26b,48d,62]

As it was proven that FeSn2/NF can be a competitive

anode for the OER, a two-electrode set-up was fabricated

to carry out the reaction of overall water-splitting using a

Pt cathode coupled to FeSn2/NF anode (Pt(−)//(+)FeSn2/

NF) in alkaline electrolyte. Remarkably, the Pt(−)//(+)FeSn2/

NF exhibited an extremely reduced cell potential of 1.53V at

10 mA cm−2, which was much lower than cell constructed with

Pt(−)//(+)RuO2/NF (1.64 V) and Pt(−)//(+)IrO2/NF (1.66 V)

(Figure5a) emphasizing the prominence of an active and con-

ductive anode for durable water electrolysis. The long-term

alkaline water electrolysis of Pt(−)//(+)FeSn2/NF was conducted

at a cell potential of 1.53V that showed notable sustainability

of the system under a longer run (Figure S50, Supporting

Information). In an inverted closed-cell (Figure S51, Supporting

Information), using Pt(−)//(+)FeSn2/NF as a pair of working

electrodes, the volumes of the evolved gases in both compart-

ments were subsequently recorded and the ratio of produced

H2 and O2 overtime was about 2:1, which matches perfectly

with the theoretically predicted water splitting ratio (Figure5b).

The Faradaic efficiency (FE) of the overall water splitting was

further calculated (in a separate experiment) using gas chroma-

tography (GC) and a FE of ≈94% was achieved for OER (FeSn2/

NF) and 97% for HER (Pt) (see details in the Supporting Infor-

mation and Table S8, Supporting Information).

Recently, the transition metal-based catalysts have gained

immense interest in the oxidation of important organic sub-

strates, including the most important substrates such as HMF,

ethanol, and acetaldehyde.[31c,e,32a] The utilization of electrooxida-

tion opens a new scope to avoid the use of toxic oxidants and

harsh reaction conditions for the electrosynthesis of value-added

organic products in a green and effective approach.[29] In addi-

tion to that, electrocatalytic alcohol oxidation has a broad context

to develop direct alcohol fuel cells (DAFCs) and it has been well-

studied with noble metal (Pt, Pd, and Au) based electrodes,[32b]

whereas, the practical application of non-noble anodes is still

limited. Inspired by the promising OER activity of FeSn2/NF,

the catalytic efficiency of this electro(pre)catalyst was used for

the electrooxidation of ethanol, carboxaldehyde, and HMF.

The electrocatalytic oxidation of ethanol was tested in

aqueous 1 m KOH using a three-electrode set-up in a similar

Figure 5. The FeSn2/NF anodes for electrocatalytic water splitting and organic oxidation. a) Polarization curves measured for Pt(−)//(+)FeSn2/NF,

Pt(−)//(+)IrO2/NF, Pt(−)//(+)RuO2/NF and NF(−)//(+)NF. b) Evolution of stoichiometric amount of H2 and O2 (2:1) during electrolysis. c) Polarization

curves recorded in a three-electrode set-up with FeSn2/NF as a working electrode in aqueous 1 m KOH in the absence and presence of 1 m ethanol

(sweep rate; 5mV s−1). d) Schematic representation of water electrolysis and electrocatalytic oxidation of alcohols with FeSn2/NF as anode and Pt as

the cathode (in a divided cell).

Adv. Energy Mater. 2020, 10, 2001377

www.advenergymat.dewww.advancedsciencenews.com

fashion to that of OER. Polarization curves obtained from the

LSV measurement showed a rapid increment in current den-

sity beyond 1.35V (vs RHE) when a mixture of 1 m KOH and

1 m ethanol was used as electrolyte solution (Figure5c). How-

ever, in the absence of 1 m ethanol in the electrolyte (only 1 m

KOH), the LSV curve showed only the OER and revealed that

the oxidation of ethanol transpired significantly lower poten-

tials than that of OER. This study suggests that the Fe0 in FeSn2

oxidizes to form higher-valent Fe species which spontaneously

oxidize ethanol in comparison to that of alkaline water. In order

to comprehend the outcome of ethanol after electrooxidation, a

customized two-compartment divided cell set-up was fabricated

using FeSn2/NF as anode and Pt as the cathode (Figure5d).

At a constant current of 10 mA cm−2, the ethanol converted

into acetic acid as the sole four-electron oxidized product

(–CH2OH to –COOH; 4 e− oxidation) at the anode, as identified

and quantified by 1H NMR (Figures S52 and S53, Supporting

Information) and at the cathode, a continuous H2 production

was observed. The achieved FE of ethanol to acetic acid con-

version (at the anode) is ≈90% after 2 h of applied current

implying a potency of the Pt(−)//(+)FeSn2/NF cell. Similar to

ethanol oxidation, the two-electrode set-up was also constructed

to electrooxidize acetaldehyde to acetic acid and ≈85% FE was

also achieved at a constant current of 10mA cm−2 (Figure S54,

Supporting Information). In fact, the acetic acid (CH3COOH),

the oxidation product of C2H5OH and CH3CHO, is a

high boiling polar solvent used in common practice for

standard organic synthesis. Besides, a paired C2H5OH oxida-

tion with simultaneous H2 formation are two important steps

that occur in direct ethanol fuel cell and FeSn2/NF anode with

90% efficiency. This system would be a potential alternative to

the noble metal anodes traditionally used to achieve the highest

efficiency.

The electrosynthesis of FDCA from HMF oxidation is a

value-added component for polymer manufacturing, and poly-

ester synthesis, for instance, polyethylene 2,5-furandicarboxy-

late (PEF) and polyethylene terephthalate.[29] HMF contains a

dual functionality; –CHO and –CH2OH and using FeSn2/NF as

the anode, electrolysis for 2 h in an alkaline solution (1 m KOH)

containing 20 × 10−3 m of HMF, FDCA was obtained as the

only oxidation product (Figures S55 and S56, Supporting Infor-

mation) with a FE of ≈90%. Interestingly, the FE for all mul-

tistep electrocatalytic oxidation reactions (Schemes S1 and S2,

Supporting Information), acetic acid from ethanol (4 e−), from

acetaldehyde (2 e−) and HMF to FDCA (6 e−), is extremely high

in comparison to the literature reports and are in the range of

85–90%, indicating superior efficiency.[36–38] This study shows

the use of ordered intermetallic FeSn2 for the first time for the

electrooxidation of organics of value-added chemicals as well as

OER.

In summary, we report a straightforward synthetic strategy

to isolate a highly crystalline intermetallic FeSn2 nanoparti-

cles, avoiding traditional high-temperature solid-state tech-

niques. While deposited on electrode substrates, FeSn2

behaves as a potential anode material for electrocatalytic

OER with a substantially low overpotential of 273 mV on

the FTO surface and 197 mV on NF (at 10 mA cm−2), sur-

passing the previously reported pure iron-based materials

for OER. The superior activity of the FeSn2 catalyst is most

likely due to higher electron conductivity within the struc-

turally ordered metallic nanostructure compared to other

iron-based catalysts reported earlier and/or tested herein, as

evident by comparatively low resistances obtained from the

four-point probe resistivity measurements. Detailed micro-

scopic and spectroscopic postanalyses indicate the formation

of a crystalline α-FeO(OH) (goethite) as an overlayer via a

surface restructuring to act as an active species for OER on

the surface of highly conducting FeSn2. Utilization of this

intermetallic FeSn2 nanostructure using Fe, one of the most

earth-abundant metals, as the active site to form the surface-

active α-FeO(OH) species and FeSn2 at the core as a conduc-

tive site is found to be beneficial compared to earlier strategies

to improve the poor conductivity of iron-based materials by

doping with one or two additional transition metal (Ni/Co)

and/or varying different conductive substrates. The superior

OER performance of this in situ generated α-FeO(OH)@

FeSn2 catalyst provides an additional opportunity to construct

a reliable overall water splitting cell Pt(−)//(+)FeSn2/NF with

a cell potential of 1.53V. We further utilized the nonprecious

α-FeO(OH)@FeSn2 anode for the electrosynthesis of two valu-

able organic compounds acetic acid and FDCA starting from

ethanol/acetaldehyde and HMF, respectively, with highest FE.

The findings obtained here are vital steps forward in the direc-

tion of the rational design of low-cost electrodes for future

electrochemical systems involving multiple applications such

as OER, water electrolysis, and green chemical synthesis.

Supporting Information

Supporting Information is available from the Wiley Online Library or

from the author.

Acknowledgements

B.C. and R.B.-S. contributed equally to this work. The work is

funded by the Deutsche Forschungsgemeinschaft (DFG, German

Research Foundation) under Germany´s Excellence Strategy—EXC

2008—390540038—UniSysCat. The authors are also indebted to Dr.

Vitaly Gutkin for XPS measurement and to Dr. Stefan Berendts, Eva

Maria Heppke, and Ina Speckmann (AK Lerch, TU Berlin) for PXRD

measurements, and Konstantin Laun (AK Hildebrandt) for Raman

measurements.

Conflict of Interest

The authors declare no conflict of interest.

Keywords

iron stannides, overall water splitting, oxygenation, oxyhydroxides, water

oxidation

Received: April 21, 2020

Revised: June 8, 2020

Published online: June 24, 2020

Adv. Energy Mater. 2020, 10, 2001377

www.advenergymat.de

www.advancedsciencenews.com

[1] a) J. Qi, W. Zhang, R. Cao, Adv. Energy Mater. 2018, 8, 1701620;

b) Y.Wang, H.Suzuki, J.Xie, O.Tomita, D. J.Martin, M.Higashi,

D.Kong, R.Abe, J.Tang, Chem. Rev. 2018, 118, 5201; c) H. B.Gray,

Nat. Chem. 2009, 1, 7; d) T. A.Faunce, W.Lubitz, A. W.Rutherford,

D. MacFarlane, G. F. Moore, P. Yang, D. G.Nocera, T. A. Moore,

D. H. Gregory, S. Fukuzumi, K. B. Yoon, F. A. Armstrong,

M. R. Wasielewski, S. Styring, Energy Environ. Sci. 2013, 6, 695;

e) C. Panda, P. W. Menezes, S. L. Yao, J. Schmidt, C. Walter,

J. N.Hausmann, M.Driess, J. Am. Chem. Soc. 2019, 141, 13306.

[2] a) T. J.Meyer, M. V.Sheridan, B. D.Sherman, Chem. Soc. Rev. 2017,

46, 6148; b) D. A.Pantazis, ACS Catal. 2018, 8, 9477; c) C.Panda,

P. W.Menezes, M.Driess, Angew. Chem., Int. Ed. 2018, 57, 11130.

[3] B.You, Y.Sun, Acc. Chem. Res. 2018, 51, 1571.

[4] a) J. H. Montoya, L. C. Seitz, P. Chakthranont, A. Vojvodic,

T. F. Jaramillo, J. K. Nørskov, Nat. Mater. 2017, 16, 70;

b) B. M. Hunter, H. B. Gray, A. M. Müller, Chem. Rev. 2016, 116,

14120.

[5] a) Y. Jiao, Y. Zheng, M. Jaroniec, S. Z. Qiao, Chem. Soc. Rev.

2015, 44, 2060; b) I. Roger, M. A. Shipman, M. D. Symes, Nat.

Rev. Chem. 2017, 1, 928; c) C. C. L.McCrory, S.Jung, I. M.Ferrer,

S. M.Chatman, J. C.Peters, T. F.Jaramillo, J. Am. Chem. Soc. 2015,

137, 4347; d) R.Gao, D.Yan, Adv. Energy Mater. 2020, 10, 1900954.

[6] a) F.Song, L.Bai, A.Moysiadou, S.Lee, C.Hu, L.Liardet, X.Hu,

J. Am. Chem. Soc. 2018, 140, 7748; b) B. Chakraborty, A. Indra,

P. V.Menezes, M.Driess, P. W.Menezes, Mater. Today Chem. 2020,

15, 100226.

[7] M. W.Kanan, Y.Surendranath, D. G.Nocera, Chem. Soc. Rev. 2009,

38, 109.

[8] a) Y. Guo, T. Park, J. W. Yi, J. Henzie, J. Kim, Z. Wang, B. Jiang,

Y.Bando, Y. Sugahara, J. Tang, Y.Yamauchi, Adv. Mater. 2019, 31,

1807134; b) S. Anantharaj, S. R. Ede, K. Sakthikumar, K. Karthick,

S.Mishra, S.Kundu, ACS Catal. 2016, 6, 8069; c) B.Chakraborty,

S. Kalra, R. Beltran-Suito, C. Das, T. Hellmann, P. W. Menezes,

M. Driess, Chem. - Asian J. 2020, 15, 852; d) B. Chakraborty,

R.Beltran-Suito, V.Hlukhyy, J.Schmidt, P. W.Menezes, M.Driess,

ChemSusChem 2020, https://doi.org/10.1002/cssc.202000445;

e) J. N. Hausmann, E. M. Heppke, R. Beltran-Suito, J. Schmidt,

M.Muhlbauer, M.Lerch, P. W.Menezes, M.Driess, ChemCatChem

2020, 12, 1161.

[9] a) K. N. Dinh, Q. Liang, C.-F. Du, J. Zhao, A. I. Y. Tok, H. Mao,

Q.Yan, Nano Today 2019, 25, 99; b) B. R.Wygant, K.Kawashima,

C. B.Mullins, ACS Energy Lett. 2018, 3, 2956.

[10] S.Gupta, M. K.Patel, A.Miotello, N.Patel, Adv. Funct. Mater. 2020,

30, 1906481.

[11] P. W. Menezes, C. Panda, S. Loos, F. Bunschei-Bruns, C. Walter,

M.Schwarze, X.Deng, H.Dau, M.Driess, Energy Environ. Sci. 2018,

11, 1287.

[12] P. W. Menezes, A. Indra, I. Zaharieva, C. Walter, S. Loos,

S. Hoffmann, R. Schlögl, H. Dau, M. Driess, Energy Environ. Sci.

2019, 12, 988.

[13] H. E.Suess, H. C.Urey, Rev. Mod. Phys. 1956, 28, 53.

[14] V.Abbate, R.Hider, Metallomics 2017, 9, 1467.

[15] a) S.Liu, L.Zheng, P.Yu, S.Han, X.Fang, Adv. Funct. Mater. 2016,

26, 3331; b) S. Han, L. Hu, Z.Liang, S. Wageh, A. A. Al-Ghamdi,

Y.Chen, X.Fang, Adv. Funct. Mater. 2014, 24, 5719.

[16] a) S.Chen, Z.Kang, X.Zhang, J.Xie, H.Wang, W.Shao, X.Zheng,

W. Yan, B. Pan, Y. Xie, ACS Cent. Sci. 2017, 3, 1221; b) L. Yang,

Z.Guo, J.Huang, Y.Xi, R.Gao, G.Su, W.Wang, L.Cao, B.Dong,

Adv. Mater. 2017, 29, 1704574; c) H. Li, P. Wen, Q. Li, C. Dun,

J. Xing, C. Lu, S. Adhikari, L. Jiang, D. L. Carroll, S. M. Geyer,

Adv. Energy Mater. 2017, 7, 1700513; d) F. Yu, H. Zhou, Z. Zhu,

J. Sun, R. He, J. Bao, S. Chen, Z. Ren, ACS Catal. 2017, 7, 2052;

e) C. Panda, P. W. Menezes, C. Walter, S. Yao, M. E. Miehlich,

V. Gutkin, K. Meyer, M. Driess, Angew. Chem., Int. Ed. 2017, 56,

10506; f) S. Yao, V. Forstner, P. W. Menezes, C. Panda, S. Mebs,

E. M. Zolnhofer, M. E. Miehlich, T. Szilvási, N. Ashok Kumar,

M. Haumann, K. Meyer, H. Grützmacher, M. Driess, Chem. Sci.

2018, 9, 8590; g) H. Bandal, K. K. Reddy, A. Chaugule, H. Kim,

J. Power Sources 2018, 395, 106; h) J. Hu, S. Li, J. Chu, S. Niu,

J. Wang, Y. Du, Z. Li, X. Han, P. Xu, ACS Catal. 2019, 9, 10705;

i) X.Zou, Y.Wu, Y.Liu, D.Liu, W.Li, L.Gu, H.Liu, P.Wang, L.Sun,

Y.Zhang, Chem 2018, 4, 1139; j) Y.Wu, M.Chen, Y.Han, H.Luo,

X. Su, M.-T. Zhang, X. Lin, J. Sun, L. Wang, L. Deng, W. Zhang,

R. Cao, Angew. Chem., Int. Ed. 2015, 54, 4870; k) S. Han, X. Hu,

J.Wang, X.Fang, Y.Zhu, Adv. Energy Mater. 2018, 8, 1800955.

[17] a) M. Arif, G. Yasin, M. Shakeel, M. A. Mushtaq, W.Ye, X. Fang,

S.Ji, D.Yan, Mater. Chem. Front. 2019, 3, 520; b) R.Gao, D.Yan,

Nano Res. 2018, 11, 1883.

[18] M.Arif, G.Yasin, L.Luo, W.Ye, M. A.Mushtaq, X.Fang, X.Xiang,

S.Ji, D.Yan, Appl. Catal., B 2020, 265, 118559.

[19] a) S. Zou, M. S. Burke, M. G. Kast, J. Fan, N. Danilovic,

S. W. Boettcher, Chem. Mater. 2015, 27, 8011; b) K. Klykin,

K. M.Rosso, V.Alexandrov, J. Phys. Chem. C 2018, 122, 16086.

[20] R. Subbaraman, D. Tripkovic, K.-C. Chang, D. Strmcnik,

A. P. Paulikas, P. Hirunsit, M. Chan, J. Greeley, V. Stamenkovic,

N. M.Markovic, Nat. Mater. 2012, 11, 550.

[21] a) A. S. Batchellor, S. W. Boettcher, ACS Catal. 2015, 5, 6680;

b) L.Trotochaud, S. L.Young, J. K.Ranney, S. W.Boettcher, J. Am.

Chem. Soc. 2014, 136, 6744; c) D.Friebel, M. W.Louie, M.Bajdich,

K. E. Sanwald, Y. Cai, A. M. Wise, M.-J. Cheng, D. Sokaras,

T.-C.Weng, R.Alonso-Mori, R. C.Davis, J. R.Bargar, J. K.Norskov,

A.Nilsson, A. T.Bell, J. Am. Chem. Soc. 2015, 137, 1305.

[22] a) C. G. Morales-Guio, L. Liardet, X. Hu, J. Am. Chem. Soc.

2016, 138, 8946; b) M. S.Burke, S.Zou, L. J.Enman, J. E.Kellon,

C. A.Gabor, E.Pledger, S. W.Boettcher, J. Phys. Chem. Lett. 2015,

6, 3737; c) M. S. Burke, L. J. Enman, A. S. Batchellor, S. Zou,

S. W.Boettcher, Chem. Mater. 2015, 27, 7549; d) L.Bai, C.-S.Hsu,

D. T. L.Alexander, H. M.Chen, X.Hu, J. Am. Chem. Soc. 2019, 141,

14190.

[23] a) C. Roy, B. Sebok, S. B. Scott, E. M. Fiordaliso, J. E. Sørensen,

A. Bodin, D. B. Trimarco, C. D. Damsgaard, P. C. K. Vesborg,

O. Hansen, I. E. L. Stephens, J. Kibsgaard, I. Chorkendorff,

Nat. Catal. 2018, 1, 820; b) X. Zhang, Y. Zhao, Y. Zhao, R. Shi,

G. I. N.Waterhouse, T.Zhang, Adv. Energy Mater. 2019, 9, 1900881;

c) W. Zhang, Y. Wu, J. Qi, M. Chen, R. Cao, Adv. Energy Mater.

2017, 7, 1602547; d) M. Görlin, P. Chernev, J. Ferreira de Araújo,

T.Reier, S.Dresp, B.Paul, R.Krähnert, H.Dau, P.Strasser, J. Am.

Chem. Soc. 2016, 138, 5603; e) F.Song, X.Hu, Nat. Commun. 2014,

5, 4477; f) K. Fominykh, J. M. Feckl, J. Sicklinger, M. Döblinger,

S.Böcklein, J.Ziegler, L.Peter, J.Rathousky, E.-W.Scheidt, T.Bein,

D.Fattakhova-Rohlfing, Adv. Funct. Mater. 2014, 24, 3123.

[24] a) H. S. Ahn, A. J. Bard, J. Am. Chem. Soc. 2016, 138, 313;

b) M. Asnavandi, Y. Yin, Y. Li, C. Sun, C. Zhao, ACS Energy Lett.

2018, 3, 1515; c) N. Li, D. K. Bediako, R. G. Hadt, D. Hayes,

T. J.Kempa, F.vonCube, D. C.Bell, L. X.Chen, D. G.Nocera, Proc.

Natl. Acad. Sci. USA 2017, 114, 1486.

[25] a) F. Song, M. M. Busch, B. Lassalle-Kaiser, C.-S. Hsu,

E.Petkucheva, M.Bensimon, H. M.Chen, C.Corminboeuf, X.Hu,

ACS Cent. Sci. 2019, 5, 558; b) K.Fominykh, P.Chernev, I.Zaharieva,

J. Sicklinger, G. Stefanic, M. Döblinger, A. Müller, A. Pokharel,

S. Böcklein, C. Scheu, T. Bein, D. Fattakhova-Rohlfing, ACS Nano

2015, 9, 5180; c) S.Lee, K.Banjac, M.Lingenfelder, X.Hu, Angew.

Chem., Int. Ed. 2019, 58, 10295; d) J. Y. C.Chen, L.Dang, H.Liang,

W.Bi, J. B.Gerken, S.Jin, E. E.Alp, S. S.Stahl, J. Am. Chem. Soc.

2015, 137, 15090.

[26] a) B. M. Hunter, J. R. Winkler, H. B. Gray, Molecules 2018, 23,

903; b) J. M. P.Martirez, E. A.Carter, J. Am. Chem. Soc. 2019, 141,

693; c) L. J. Enman, M. B. Stevens, M. H. Dahan, M. R. Nellist,

M. C. Toroker, S. W. Boettcher, Angew. Chem., Int. Ed. 2018, 57,

12840.

Adv. Energy Mater. 2020, 10, 2001377

www.advenergymat.dewww.advancedsciencenews.com

[27] Q.Shi, Q.Liu, Y.Ma, Z.Fang, Z.Liang, G.Shao, B.Tang, W.Yang,

L.Qin, X.Fang, Adv. Energy Mater. 2020, 10, 1903854.

[28] Y. Kwon, K. J. P. Schouten, J. C. van der Waal, E. de Jong,

M. T. M.Koper, ACS Catal. 2016, 6, 6704.

[29] a) A. J. J. E.Eerhart, A. P. C.Faaij, M. K.Patel, Energy Environ. Sci.

2012, 5, 6407; b) R.-J. van Putten, J. C. van der Waal, E. de Jong,

C. B.Rasrendra, H. J.Heeres, J. G.deVries, Chem. Rev. 2013, 113,

1499.

[30] K. R.Vuyyuru, P.Strasser, Catal. Today 2012, 195, 144.

[31] a) P. Zhang, X. Sheng, X. Chen, Z. Fang, J.Jiang, M. Wang, F. Li,

L.Fan, Y.Ren, B.Zhang, B. J. J.Timmer, M. S. G.Ahlquist, L.Sun,

Angew. Chem., Int. Ed. 2019, 58, 9155; b) N.Zhang, Y.Zou, L.Tao,

W. Chen, L. Zhou, Z. Liu, B. Zhou, G. Huang, H. Lin, S. Wang,

Angew. Chem., Int. Ed. 2019, 58, 15895; c) S. Barwe, J. Weidner,

S. Cychy, D. M. Morales, S. Dieckhöfer, D. Hiltrop, J. Masa,

M.Muhler, W.Schuhmann, Angew. Chem., Int. Ed. 2018, 57, 11460;

d) W. Li, N. Jiang, B. Hu, X. Liu, F. Song, G. Han, T. J. Jordan,

T. B. Hanson, T. L. Liu, Y. Sun, Chem 2018, 4, 637; e) W.-J. Liu,

L.Dang, Z.Xu, H.-Q.Yu, S.Jin, G. W.Huber, ACS Catal. 2018, 8,

5533.

[32] a) J. Bai, D. Liu, J. Yang, Y. Chen, ChemSusChem 2019, 12, 2117;

b) Y.Wang, S.Zou, W.-B.Cai, Catalysis 2015, 5, 1507.

[33] R. J.Cava, H.Takagi, H. W.Zandbergen, J. J.Krajewski, W. F.Peck,

T.Siegrist, B.Batlogg, R. B.vanDover, R. J.Felder, K.Mizuhashi,

J. O.Lee, H.Eisaki, S.Uchida, Nature 1994, 367, 252.

[34] M.Armbrüster, R.Schlögl, Y.Grin, Sci. Technol. Adv. Mater. 2014, 15,

034803.

[35] a) E. Edison, W. C. Ling, V. Aravindan, S. Madhavi, ChemE-

lectroChem 2017, 4, 1932; b) S. Wang, M. He, M. Walter,

F.Krumeich, K. V.Kravchyk, M. V.Kovalenko, Nanoscale 2018,

10, 6827.

[36] F.Song, K.Schenk, X.Hu, Energy Environ. Sci. 2016, 9, 473.

[37] P. W.Menezes, C.Panda, S.Garai, C.Walter, A.Guiet, M.Driess,

Angew. Chem., Int. Ed. 2018, 57, 15237.

[38] K. Karthick, S. Anantharaj, S. Patchaiammal, S. N. Jagadeesan,

P.Kumar, S. R.Ede, D. K.Pattanayak, S.Kundu, Inorg. Chem. 2019,

58, 9490.

[39] F. T.Seitz, D.Turnbull, H.Ehrenreich, Solid State Physics, Academic

Press, New York 1968.

[40] N. H.Chou, R. E.Schaak, J. Am. Chem. Soc. 2007, 129, 7339.

[41] M.Armbrüster, W.Schnelle, R.Cardoso-Gil, Y.Grin, Chem. - Eur. J.

2010, 16, 10357.

[42] M.Armbrüster, M.Schmidt, R.Cardoso-Gil, H.Borrmann, Y.Grin,

Z. Kristallogr. - New Cryst. Struct. 2007, 222, 83.

[43] Y. Yin, R. M. Rioux, C. K. Erdonmez, S. Hughes, G. A. Somorjai,

A. P.Alivisatos, Science 2004, 304, 711.

[44] a) A. T. Brown, K. G. Barron, B. G. Salazar, P. Kirby,

G. T. McCandless, A. V. Walker, J. Y. Chan, Z. Anorg. Allg. Chem.

2017, 643, 2038; b) C.Wang, J.Wang, H. Chen, M. Wen, K.Xing,

S. Chen, Q. Wu, Sci. China Mater. 2016, 59, 927; c) C. Leostean,

O. Pana, M. Stefan, A. Popa, D. Toloman, M. Senila, S. Gutoiu,

S.Macavei, Appl. Surf. Sci. 2018, 427, 192.

[45] K. K. D.Ehinon, S.Naille, R.Dedryvère, P.-E.Lippens, J.-C.Jumas,

D.Gonbeau, Chem. Mater. 2008, 20, 5388.

[46] P.Li, X.Duan, Y.Kuang, Y.Li, G.Zhang, W.Liu, X.Sun, Adv. Energy

Mater. 2018, 8, 1703341.

[47] J.Kibsgaard, I.Chorkendorff, Nat. Energy 2019, 4, 430.

[48] a) J.Masud, S.Umapathi, N.Ashokaan, M.Nath, J. Mater. Chem. A

2016, 4, 9750; b) D.Zhong, L.Liu, D.Li, C.Wei, Q.Wang, G.Hao,

Q. Zhao, J. Li, J. Mater. Chem. A 2017, 5, 18627; c) C. Zhang,

B. Zhang, Z. Li, J. Hao, ACS Appl. Energy Mater. 2019, 2, 3343;

d) J. Zou, G. Peleckis, C.-Y. Lee, G. G. Wallace, Chem. Commun.

2019, 55, 8808; e) M. Abu Sayeed, G. J. Millar, A. P. O’Mullane,

ChemElectroChem 2019, 6, 3667.

[49] C.Feng, M. B.Faheem, J.Fu, Y. Q.Xiao, C. L.Li, Y. B.Li, ACS Catal.

2020, 10, 4019.

[50] a) Q. Kang, L. Vernisse, R. C. Remsing, A. C. Thenuwara,

S. L.Shumlas, I. G.McKendry, M. L.Klein, E.Borguet, M. J.Zdilla,

D. R.Strongin, J. Am. Chem. Soc. 2017, 139, 1863; b) Y.Yoon, B.Yan,

Y.Surendranath, J. Am. Chem. Soc. 2018, 140, 2397.

[51] a) P. W. Menezes, C. Walter, J. N. Hausmann, R. Beltran-Suito,

C. Schlesiger, S. Praetz, V. Y. Verchenko, A. Shevelkov, M. Driess,

Angew. Chem., Int. Ed. 2019, 58, 16569; b) C.Panda, P. W.Menezes,

M.Zheng, S.Orthmann, M.Driess, ACS Energy Lett. 2019, 4, 747;

c) X.Xu, F.Song, X.Hu, Nat. Commun. 2016, 7, 12324.

[52] a) D. L. A. de Faria, F. N. Lopes, Vib. Spectrosc. 2007, 45, 117;

b) D. L. A.deFaria, S.Venâncio Silva, M. T.deOliveira, J. Raman

Spectrosc. 1997, 28, 873; c) W. Luo, C. Jiang, Y. Li, S. A. Shevlin,

X. Han, K. Qiu, Y. Cheng, Z. Guo, W. Huang, J. Tang, J. Mater.

Chem. A 2017, 5, 2021.

[53] a) M. B. Stevens, L. J. Enman, A. S. Batchellor, M. R. Cosby,

A. E. Vise, C. D. M. Trang, S. W. Boettcher, Chem. Mater. 2017,

29, 120; b) P.Yu, Z.Zhang, L.Zheng, F.Teng, L.Hu, X.Fang, Adv.

Energy Mater. 2016, 6, 1601111.

[54] S.Klaus, L.Trotochaud, M.-J.Cheng, M.Head-Gordon, A. T.Bell,

ChemElectroChem 2016, 3, 66.

[55] R. D. L.Smith, M. S.Prévot, R. D.Fagan, Z.Zhang, P. A.Sedach,

M. K. J.Siu, S.Trudel, C. P.Berlinguette, Science 2013, 340, 60.

[56] W.Ye, Y.Yang, X.Fang, M.Arif, X.Chen, D.Yan, ACS Sustainable

Chem. Eng. 2019, 7, 18085.

[57] R.Gao, H.Zhang, D.Yan, Nano Energy 2017, 31, 90.

[58] C. C. L.McCrory, S.Jung, J. C.Peters, T. F.Jaramillo, J. Am. Chem.

Soc. 2013, 135, 16977.

[59] a) P. W.Menezes, A.Indra, C.Das, C.Walter, C.Göbel, V.Gutkin,

D.Schmeiβer, M.Driess, ACS Catal. 2017, 7, 103; b) P. W.Menezes,

C. Panda, C. Walter, M. Schwarze, M. Driess, Adv. Funct. Mater.

2019, 29, 1808632.

[60] a) T. Wu, S. Sun, J. Song, S. Xi, Y. Du, B. Chen, W. A.Sasangka,

H. Liao, C. L. Gan, G. G. Scherer, L. Zeng, H. Wang, H. Li,

A.Grimaud, Z. J.Xu, Nat. Catal. 2019, 2, 763; b) Z. Wu, Q.Gan,

X.Li, Y.Zhong, H.Wang, J. Phys. Chem. C 2018, 122, 2848; c) Z.Wu,

X.Li, W.Liu, Y.Zhong, Q.Gan, X.Li, H.Wang, ACS Catal. 2017, 7,

4026; d) B.Yan, D.Krishnamurthy, C. H. Hendon, S. Deshpande,

Y.Surendranath, V.Viswanathan, Joule 2017, 1, 600.

[61] D. Fu, P. G. Keech, X. Sun, J. C. Wren, Phys. Chem. Chem. Phys.

2011, 13, 18523.

[62] Q.Xiang, F.Li, W.Chen, Y.Ma, Y.Wu, X.Gu, Y.Qin, P.Tao, C.Song,

W.Shang, H.Zhu, T.Deng, J.Wu, ACS Energy Lett. 2018, 3, 2357.

Adv. Energy Mater. 2020, 10, 2001377