The family of transition metal chalcogenides has attracted tremendous attention in the materials research community due to its promising future in sensors, energy conversion and energy storage among other applications [1e5]. In addition, transition metal based electrocatalysts for electrochemical energy conversion has attracted significant attention during last several years owing to their high activity, low cost, and facile tunability. Among these electrocatalytic water splitting which includes hydrogen evolution and oxygen evolution reactions (HER and OER, respectively) has been the focus of major research activity [1,6]. The challenge for practical implementation of water splitting technology, however, depends on identification of suitable and stable catalysts, which can lower reaction energy barrier and increase Faraday efficiency for both reactions. Traditionally, Pt metal is most active as HER catalyst, whereas oxides of iridium (Ir), and ruthenium (Ru), exhibit higher OER performance. However, high price and low stability of these precious metal-based catalysts make commercialization difficult. Transition metal-based chalcogenides (TMCs) have recently shown unprecedented high efficiency for HER and OER in wide range of pH. Apart from water splitting, recently TMCs have also shown tremendous success in electrocatalytic CO 2 reduction (CO2RR). Continuous dependence on fossil fuels for several decades had led to accumulation of high levels of CO 2 , which has led to several catastrophic environmental effects threatening well-being of mankind. One of the prime objectives of researchers is to develop technologies that can convert CO 2 to useful products thereby closing the carbon loop. Transition metal-based compounds have been specially attractive for such CO 2 utilization technologies, among which the chalcogenides offer special advantage owing to their selectivity towards forming value-added carbonrich reduction products. [7e11] In this perspective we will highlight recent progress made in transition metal chalcogenides based electrocatalysts for water splitting and CO2RR, and discuss the reasons behind their high activity (Figure 1).
Two major half-reactions, HER and OER, are responsible for total water splitting to create pure H 2 and O 2 . Although thermodynamic water splitting voltage is 1.23 V, a much larger potential is required due to energy loss in the electrochemical system necessitating use of suitable electrocatalysts [12]. Several TMCs have recently demonstrated promising performance for HER and OER, surpassing noble metal-based catalysts [1,13,14]. The TMCs typically exhibit low intrinsic electrical resistivity and rapid charge transfer owing to their electron rich transition metal centers, highly covalent network, and electrochemical tunability. These TMCs are primarily comprised of selenides and tellurides of cobalt, nickel, iron, and copper, in binary and multinary compounds. By modifying the intrinsic and extrinsic structures and compositions of TMCs, a variety of compounds containing different transition metals and chalcogens can be synthesized. In the following sections, binary and multinary (ternary, quaternary, and more complex) chalcogenides have been discussed with respect to electrocatalytic water splitting.
Binary metal chalcogenides, more commonly referred to as TMCs, contain one type of transition metal and one type of chalcogen. The most well-known example among them is nickel based binary metal chalcogenides (Ni x Se y ), which has been widely studied for water splitting. A variety of stoichiometric (Nise, NiSe 2 , Ni 3 Se 2 , Ni 3 Se 4 ) and non-stoichiometric (Ni 0.85 Se)
Since the electron density around the catalytically active transition metal center plays a critical role in defining catalytic activity, it was impervious to tune that electron density through structural and compositional modifications. While changing anions in the chalcogenide series leads to change in lattice electronegativity, doping in the transition metal site can lead to more subtle change in electron density around the catalytic center through ded transitions and advent of metalemetal bonding. Hence, mixed-transition-metal based electrocatalysts have emerged as new pH-universal electrocatalysts for watersplitting, which has also been characterized by their enhanced electrical conductivity, synergistic impact of bimetallic compositions, and structural stability. Several multinary TMC composites with tunable compositions were investigated in addition to cobalt and nickel chalcogenides. These composites typically outperformed their monometallic counterparts in terms of catalytic performance as has been exemplified through several examples [20,21]. As an example, superior conductivity and reactive properties of NieCo bimetallic selenides have been identified as viable electrocatalysts for OER [21]. NiCoSe 2 @PCM showed outstanding catalytic activity as a bifunctional electrocatalyst, with overpotentials of 116 mV and 140 mV at 10 mA cm À2 for HER and OER, respectively [ The catalytic performance of these MeNiSe/NF materials is M-dependent, and it follows the order VeNiSe > CoeNiSe > CueNiSe > NiSe toward the OER and CueNiSe > CoeNiSe > NiSe > VeNiSe toward the HER. Doping of NiSe with different transition metals is an example of intrinsic effect. An intrinsic doping consisting of doping in the transition metal site, leads to scrambling of the active sites and modulation of electron density around the active site leading to change in catalytic activity [23]. Although this review is focused on alkaline water electrolysis, it must be noted that water splitting has also been conducted in acidic electrolyte to be more compatible with fuel cell systems. Accordingly electrocatalysts for acid water electrolysis has received substantial attention because of its high reaction efficiency, high current density, low overpotential, and compact design [28e30]. However, the functional and compositional instability of the chalcogenide based electrocatalysts in acidic medium over extended period of time limits their practical application. The overpotentials of the various metal chalcogenide catalysts with high OER activity reported by different research groups has been benchmarked in terms of the overpotential at 10 mA cm À2 and shown in Figure 2(k). Analysis of this benchmarking figure confirms certain trends: (i) decreasing electronegativity in the chalcogenide series leads to decreasing overpotential with tellurides forming the most efficient OER electrocatalysts; (ii) transition metal doping enhances electrocatalytic activity for OER making ternary and quaternary chalcogenides as better catalysts than the binaries in the same series. However, for such aliovalent substitution, doping with transition metals with fewer d-electrons is found to be more beneficial. This outlook can now be applied for better catalyst design incorporating less electronegative anions along with aliovalent substitution in the catalytically active site. Table 1 represents the progress in binary and multinary metal chalcogenides as bifunctional electrocatalysts for overall water splitting.
Research on electrocatalytic conversion of CO 2 has evolved over past few years with respect to developing catalyst composition with special emphasis on achieving selectivity for the reduction products [7e11]. In the following section we will discuss first about Cu and other transition metal-based catalyst and then focus on transition metal chalcogenides. We will highlight new CO2RR catalysts, which will provide fundamental understanding of catalytic processes to achieve product selectivity, that can be helpful for development of efficient CO2RR catalysts in future. From early on transition-based electrocatalysts have been proved as efficient candidates for CO2RR in commercial devices [80]. They have the ability to convert CO 2 into variety of hydrocarbons with good faradaic efficiency. While CO 2 conversion is of utmost importance to reduce atmospheric levels of this greenhouse gas, production of carbon monoxide, i.e., CO as the reduction product does not really mitigate the problem, since CO itself is a toxic gas. The produced CO, hence, needs to be further processed to other carbonaceous product with higher value. Ongoing research is based on improving selectivity for value-added carbon-rich products. It was hypothesized that adsorption energy of intermediate *CO on the transition metal site played a critical role in influencing product composition and selectivity [7]. Specifically, catalysts with smaller adsorption *CO adsorption energy led to preferential formation of CO and formic acid as the reduction product (Figure 3(a)).
Catalysts with significantly larger *CO adsorption energy, on the other hand, exhibited surface poisoning with CO. Hence, recent efforts have been guided towards designing catalyst compositions that will have *CO adsorption energy in the moderate range as shown in Figure 3(a). Since metal-to-ligand back bonding plays a major role in deciding binding of intermediate *CO to the transition metal site, optimizing *CO adsorption energy has been attempted by modulating d-electron density near the transition metal site through investigating different transition metals and changing anion coordination. The following sections compile various CO2RR catalysts that has been reported based on their d-electron density, and also highlights the correlation of *CO adsorption energy with product selectivity as has been studied in chalcogenide-based catalysts.
Copper was found to have enhanced activity towards CO2RR producing at least 16 different C1eC3 hydrocarbon/oxygenate products, which is better than other metals but the selectivity was not attained towards a specific product [81]. Other transition metals such as Au [82e86], Ag [81,87,88], Pd [89], Co [11,90], Zn [11, 91,92], led to formation of CO as the major product of CO 2 reduction. Similarly, other metal catalysts such as Pb [93], In [94], Sn [89,95], and Bi [96,97], also led to formation of CO predominantly. Polycrystalline Ag is known for producing CO with Faradaic efficiency (FE) of 22.4% at À0.8 V RHE [11]. DFT based studies revealed that most of these metal catalysts had a weak binding energy towards CO, which led to weaker adsorption and formation of CO as major product. Therefore, there was a need to modify transition metals such as Cu and other metals to get better selectivity for C2þ products.
Copper chalcogenides and other transition metals with d 5d 10 electrons It has been proposed that chalcogenide anions increased lattice covalency owing to their decreased electronegativity, which led to enrichment of d-electron density near the transition metal site [16]. Hence, it was expected that chalcogenides will show better *CO adsorption energy leading to better selectivity of C2 products [7, 91,98]. Indeed DFT calculation revealed that Cu-selenides showed better *CO adsorption energy compared to the base metal and its oxide as shown in Figure 3(a). Experimental studies also confirm that chalcogens favor the formation of C2 products and deter the formation of C1 products because of decrease in surface concentration of CO [11,91]. Investigation of various Cu-based catalysts such as Cu, Cu-oxide, Cusulfide and Cu-selenide catalysts, revealed that selectivity towards certain carbon products increased with Cu-chalcogenides compared to Cu. This fact can be proven by comparing polycrystalline Cu [91,99], and CuS having HCOOH product with FE = 80% at À0.8
V RHE [91], 3D-Cu 2 S showing improved performance with a FE = 87.3% at À0.9 V RHE [100], while, Cu 2 Se nanocubes [7], and Cu 1. 63 Se nanowires [101], Cu 1.8 Se nanowires [11], showed a high C2 product selectivity. In case of other transition metals also chalcogenides showed improved activity and selectivity.
FeS 2 /NiS showed very good performance in reduction of CO 2 to form CH 3 OH with overpotential of 280 mV and Faradaic efficiency of 64% [102]. Using molecular engineering, iron porphyrin was synthesized, which showed good activity achieving 98% Faradaic efficiency for CO production [11]. Cobalt Phthalocyanine with pyridyl moieties as axial ligands achieved 70% FE towards CO production [82]. This study showed that upon axial coordination to CoPc, catalytic activity towards CO2RR improved due to energy level rise of d z 2 orbital [103]. When Ag was modified by sulfur to form Ag 2 S, selectivity improved to 92.0% at very low overpotential (À0.754 V) [82,104]. Zn has shown limited activity towards CO2RR [105]. Zn nanosheets exhibit Faradaic efficiency of 70.9% when producing CO from CO 2 electrochemical reduction. Upon modification with sulfur to make ZnS/Zn, the FE increased to 94.2% [106]. ZnTe is also an efficient catalyst to photochemically convert CO 2 [107]. Similarly Cd also changes selectivity from formic acid to CO when it is converted to CdS [74]. CdS showed good stability of 40þ hours and 95% FE towards CO during CO2RR [11]. Sn is also known to be an active metal towards CO2RR often modulated by surface structure. Modifications such as change in particle size [108], oxide layer thickness [109], morphology [110], and electrolyte's pH have improved the activity, selectivity, energetic efficiency, and robustness of Sn-based catalysts [111,112]. As reported in the study, tin oxide (SnO x ) layer was found to be active because CO 2 intermediate adheres better to the catalyst surface, and oxide layer suppresses the competing hydrogen evolution reaction [113,114]. The catalytic efficiency was improved further by converting SnO x to SnSe 2 , which balances the binding strength of CO 2 intermediates and also improves conductivity as compared to oxide layer [115]. The electronic conductivity of metal selenides was found to be better due to higher electronegativity of oxygen than selenium. This also leads to the fast charge transfer during catalytic activity [115]. Hence, SnSe 2 showed high current density (12.0 mA cm À2 ) and high FE (88.4%) at very low potential (À0.76 V vs RHE) for CO2RR [115].
Transition metal with d 1 -d 3 electrons Liu et al. showed that ultrathin MoTe 2 layers can reduce CO 2 to CH 4 with faradaic efficiency of 83% with extended stability of 45 h. Ultrathin layers lead to highly efficient mass transport due to layered structure, which makes active sites available and improves CO 2 electrochemical reduction [116]. To improve conductivity and number of active sites two-dimensional (2D) MoSe 2 doped with transition metals (Fe, Co, Ni, and Cu) were explored using density functional theory (DFT) calculations. Cu doped MoSe 2 was found to potentially have great activity for methane production [117]. WSe 2 nanoflakes show activity towards CO2RR at very low overpotential of 0.054 V forming CO as main product (FE = 24%) [82].
Transition metal chalcogenides have gained considerable attraction in the last few years as electrocatalysts for various electrochemical energy conversion processes owing to their tunable electrochemical properties, structural richness, enhanced charge transfer ability, and possibility of tuning electron density at the catalytic center through doing and vacancy ordering. In this review we have highlighted their high activity for electrocatalytic water splitting and CO 2 electroreduction. Various binary and multinary metal chalcogenides catalysts have been reported for OER with significantly improved catalytic performance compared to the state-of-the-art precious metal-based oxides as shown in Figure 2k. In particular, these catalysts exhibited low h 10 overpotentials primarily due to facile catalyst activation through enhanced electrochemical tunability of the transition metal center, which is facilitated by the lower electronegativity and higher covalency of the chalcogenide anions. These semiconducting TMCs also have a high potential for use in solar water splitting due to their ideal band gap and acceptable band-edge positions that align well with both water reduction and oxidation potentials. Transition metal dichalcogenides (TMDCs) were previously investigated as potential candidates for photoelectrodes with efficiencies of up to 17% when illuminated in acidic solutions utilizing n-type single crystals [28e30]. Coupling these transition metal chalcogenide electrocatalysts with efficient photo-absorbers to create a hybrid module will lead to significant advances in solar water splitting, by combining low overpotential and functional stability of the electrocatalysts with efficient solar energy capture.
Interestingly, the transition metal chalcogenides function as highly efficient electrocatalysts for CO 2 electroreduction, producing carbon-rich products with high selectivity. Such capability makes these TMC highly applicable for CO 2 utilization, which has become an issue of prime interest to reduce carbon footprint of industrial processes. In this review, we have discussed the fundamental issues related to formation of carbonrich product from CO 2 electroreduction and its relationship with the underlying chemical reactions on the catalyst surface. By comparing the product formation on various catalyst surfaces it was observed that *CO dwell time on the surface can be used as an appropriate descriptor for tuning the product composition. The *CO dwell time depends on the CO adsorption energy on the catalytic site. Comparison of the CO adsorption energy on various catalyst surfaces showed that weak adsorption energy (leading to less *CO dwell time) resulted in C1 products, whereas very high adsorption energy leads to indefinite stay of *CO on the surface and catalyst poisoning as seen in Ni, Pd, and Pt. This observation offers insight into designing an optimal catalyst surface where the CO adsorption energy should be in the middle range as shown in Figure 3(a) to yield predominantly C2þ products. Tuning anion electronegativity and changing the transition metal center can lead to such optimal *CO adsorption energy as is shown in the benchmark Figure 3(c). The section marked as (i) and (ii) in Figure 3(c) shows catalysts producing mainly C1 products and H 2 , wherein, (i) represents the base metals, while (ii) contains the binary and ternary compounds. This can be attributed to lower *CO adsorption energy. Section (iii) represents catalysts that can produce C2þ products. Hence, future outlook for effective CO 2 reduction electrocatalysts with high selectivity for carbon-rich product involves designing the catalyst surface that can facilitate metaleligand (CO) backbonding by aligning the metal d-orbitals with the ligand (CO) p* orbitals, leading to better CO dwell time on the surface and optimal *CO adsorption energy. Such catalysts can have significant effect on CO 2 utilization by forming value-added chemicals and fuels from waste CO 2 .
Several of these TMCs have shown efficient catalytic activity for both anodic and cathodic processes such as OER-HER and OER-CO2RR, respectively, making them suitable as bifunctional electrocatalyst. Among these, the OER-CO2RR bifunctional activity is useful, since such electrocatalytic cell (as shown in Figure 3 (b)), can essentially reduce atmospheric CO 2 to value-added carbon-rich products while enriching O 2 in the atmosphere [7]. Such electrochemical setup has also been used to carry out more economical oxidation processes such as methanol oxidation, ethanol oxidation, urea oxidation etc. [125,126] It will be interesting to explore such electrocatalytic activity for TMCs. Such bifunctional activity can be lucrative for practical industrial applications since it has minimum energy expense and the total cell potential for this methanol oxidation/OER-CO2RR electrolytic cell is lower than combination of individual processes.
Papers of particular interest, published within the period of review, have been highlighted as:
Nath M, De Silva U, Singh H, Perkins M, Liyanage WPR, Umapathi S, Chakravarty S, Masud J: Cobalt telluride: a highly efficient trifunctional electrocatalyst for water splitting and oxygen reduction. ACS Appl Energy Mater 2021:1c01438, https://doi.org/10.1021/ACSAEM.1C01438. acsaem. In an alkaline medium, cobalt telluride has been discovered to be an efficient multifunctional electrocatalyst for oxygen and hydrogen evolution reactions as well as oxygen reduction reactions. Electrodeposited CoTe demonstrated the best electrocatalytic activity for OER, requiring an overpotential of 200 mV at 10 mA cm -2 . It was observed that -OH adsorption energy on the catalytic site estimated through DFT calculations, can be used as a major descriptor for OER catalytic activity.
Singh H, Bernabe J, Chern J, Nath M: Copper selenide as multifunctional non-enzymatic glucose and dopamine sensor. , and NiS 2 ) nanocrystals have been synthesized through a simple polyol solution method and their electrocatalytic characteristics toward OER and HER have been thoroughly examined. The electrochemical results show that Ni 3 S 2 outperforms NiS and NiS 2 in terms of OER and HER performance, producing 1.63 V to achieve a current density of 10 mA cm -2 in the total water splitting device, which is comparable to that of noble metal catalysts. As reported in this paper, the performance of the as-prepared electrodes (NiSe-TMEDA/CC) in alkaline solution for water splitting was investigated. The electrodes demonstrated bifunctional OER and HER activity with small overpotentials of 226 and 30.8 mV, respectively, to achieve current densities of 100 mA cm -2 .
Current Opinion in Electrochemistry 2022, 34:100993 www.sciencedirect.com