Cobalt−tetracyanoquinodimethane monolayer as efficient Bi-functional single atom Electrocatalyst: A First Principles Investigation

A bi-functional electrocatalyst is a stable and catalytically active material for both oxygen reduction reaction (ORR) and oxygen evolution reaction (OER), the development of which is highly challenging, although essential for application in fuel cells. Herein, the first principle based density functional theory (DFT) computations have been carried out to investigate a series of transition metal-tetracyanoquinodimethane (TM-TCNQ: TM = Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Ru, Rh, Pd) monolayer for bi-functional single atom electrocatalyst towards ORR and OER. Amongst various first transition metal row based TCNQ substrates, Cobalt-TCNQ (Co-TCNQ) monolayer was predicted to exhibit best bi-functional catalytic activity. At equilibrium potential (1.23 V, vs RHE), the free energy portraits of Co-TCNQ monolayer for OER in acidic or alkaline media are not completely downhill, measuring an excess potential of 0.6 V, equal to the overpotential, which is needed to be supplied externally to promote the OER activity. On the other hand, an onset potentials of 0.34 V ( vs RHE) in acidic media are measured for ORR to proceed. The promising catalytic activity of two dimensional Co-TCNQ monolayer towards ORR and OER as revealed by DFT investigations has been explained in terms of density of states and Bader charge analysis.


Introduction
The consumption of fossil fuel in highly alarming rate and the increasing demand for energy has forced the global researcher to explore the viable alternatives in order to mitigate the crisis in energy. In recent past, fuel cells have been introduced as the novel energy conversion devices [1,2], capable of producing sustained energy through two redox reactions, namely, oxygen reduction reaction (ORR) and oxygen evolution reaction (OER). However, the main bottleneck of ORR being its sluggish reaction kinetics at cathode, which greatly limits the large scale application of fuel cells in practice. Additionally, the platinum (Pt), which has been used so far as highly effective ORR catalyst [3][4][5][6], suffers from a number of shortcomings, including, lower abundance, higher cost, and poor durability, which fatally inhibit its wide commercialization [7]. Therefore, the development of non-precious, earth-abundant and non-Pt based ORR electrocatalysts with low overpotential is highly essential for further progress of fuel cell technology. On the other hand, in case of OER, the noble metal oxides, like, ruthenium (Ru) and iridium (Ir) were identified so far to be the best catalyst [8,9], which are not only highly precious, 3 but, scarce too. This hinders their widespread scalable applications. Thus, the searching for costeffective and earth abundant electrocatalyst with comparable catalytic performances has become the urgent need. Specially, the designing of stable bi-functional electrocatalyst (ORR and OER) will be a key step towards solution of future energy catastrophe and for development of fuel cell technologies [10].
In recent past, the catalysts based on transition metal (TM)-cored macrocyclic compounds, TM anchored phthalocyanines materials and TM oxides have attracted immense research focus with promising performances [11][12][13][14][15][16][17][18][19][20]. The sheet-like structures of these 2-D metal-organic framework offers high electrical conductivity, promoting easy electron transport though the active material and high activity towards single atom electrocatalysts can be obtained through strong anchoring of the metallic centre into the pores [21]. Moreover, the substitution of central metal atom with wide variety of transition metals can modify the molecular structure which can easily tailor the electrochemical functionality of the active catalyst.
Another family of 2-D transition metal-organic molecules, viz, TMtetracyanoquinodimethane (TM-TCNQ), which is similar to the TM-macrocyclic structure, have been explored recently by different groups due to their promising photoelectrochemical and catalytic properties [22][23][24][25][26][27]. However, the span of diverse activities and the potential of these TM-TCNQ composition till remains unexplored, especially their possible extent as both ORR and OER (bi-functional) electrocatalyst is largely unknown [26]. Most of the earlier studies on various TM-TCNQ framework focussed on the feasibility of these compounds to be applied either as ORR or OER catalyst, which recognised Fe-TCNQ and Ni-TCNQ to be the supreme candidates for ORR and OER respectively, amongst other TM-TCNQ substrates [25][26][27].
However, the identification of bi-functional catalyst among undoped TM-TCNQ structures so far 4 is greatly limited. Earlier, cobalt based composite materials or cobalt doped complex structures have been reported to exhibit promising performances as bi-functional electrocatalysts [28,29].
Experimental investigation on Co-TCNQ complex revealed excellent catalytic activity with overpotential as low as 0.31 V and long term durability [30]. Also, Co doped graphene substrate in alkaline environment exhibited extremely low overpotential, ca. 0.21 V, towards OER activity [31], whereas, (Co(OH)2) nanocrystals doped with reduced graphene oxide showed an OER overpotential, ca. 0.37 V, which is lower than that of the commercial RuO2 catalyst [32].
Inspired by these reports, density functional calculation has been carried out systematically on various first and second transition metal (TM) row based TCNQ monolayer for applications towards ORR and OER in both acidic and alkaline media.
In this study, first principle calculations on two dimensional transition metaltetracyanoquinodimethane (TM-TCNQ; TM = Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Ru, Rh, Pd) monolayer were done to investigate their potential application as bi-functional electrocatalyst.
Density functional theory (DFT) computations revealed that amongst various TM−TCNQ monolayer, the best performance has been demonstrated by Co-TCNQ substrate surpassing that of the other TM-TCNQ substrates. The Co-TCNQ monolayer revealed overpotential ( ) value of as low as ca. 0.6 V (in both acidic and alkaline media) and onset potential of 0.34 V (vs RHE) in acidic media (pH = 0) towards OER and ORR, respectively. The value of for all the TM-TCNQ monolayer was found to be independent of electrolyte, i.e., the same value of was obtained in both acidic and alkaline media. DFT computations on Co-TCNQ substrate suggest that strong hybridization between Co-3d with O-2p orbitals occurred when O2 (OH * ) molecule is adsorbed on the Co-TCNQ substrate during ORR (OER). The Bader charge analysis dictated that a net charge of 0.27 | | (0.22|e|) has been transferred from Co atom centre towards O2 atom 5 (OH * ) in the Co-TCNQ-OO * (Co-TCNQ-OH * ) composite. The promising electrocatalytic performances of Co-TCNQ monolayer shows the potential to overcome the challenges towards the development of cost-effective, earth-abundant, durable and high-efficiency single-atom bifunctional catalyst for the progress of fuel cell technologies as well as for the mitigation of future energy crisis.

All computations have been carried out with Vienna Ab-initio Simulation Package (VASP)
coupled with Projector Augmented Wave (PAW) method. The generalized gradient-corrected functional with Perdew−Burke−Ernzerhof (PBE)-type exchange-correlation potential [33] have been employed in this study. The total energy has been computed using Grimme's DFT-D2 dispersion interactions which includes also Van der Waals correction [34]. For molecular geometry optimization, a 5×5×1 mesh and for electronic structure calculations, a 11×11×1 grid have been used. The Bader method was used for the calculation of charge transfer between active sites and the adsorbates. The unit cell of each TM-TCNQ monolayer consists of one metal atom at the centre of the layer, four nitrogen (N) atoms, four hydrogen (H) atoms and twelve carbon (C) atoms. The monolayer was taken parallel to the x-y plane and a vacuum layer of 20 Å thickness is set along z-direction between successive TM-TCNQ layers to discard the interlayer interactions. The energy cut-off in the projector augmented wave for the simulation and structure relaxation was set at 550 eV and the convergence threshold for energy and force in the geometry optimization were set respectively as 10 -5 eV and 0.02 eV/Å. 6

Results and Discussions
The four electron (4 − ) reduction pathway of oxygen (O2) consists of adsorption of intermediate products, viz, OOH * , O * and OH * followed by the desorption of OH − . The associative ORR mechanism in acidic environment (pH = 0) proceeds through the following equations: 2 ( ) + * → 2 * 2 * + + + − → * * + + + − → * + 2 ( ) * + + + − → * * + + + − → 2 ( ) + *  Co-TCNQ monolayer following the above reaction pathway. The adsorption energy ( ) of the reaction intermediates on Co-TCNQ surface, summarized in Table 1, has been calculated from the following equation: where, Co−TCNQ−adsorbate is the energy of the composite system, Co−TCNQ is the energy of the  (Figure 1b). This indicates the possible transfer of electronic charges from central Co-atom to the adsorbed O-atoms in the composite and initiation of ORR mechanism. In the first hydrogenation process, the activated OO * adsorb a proton (H + ) coupled    To investigate about OER, which is the reverse reaction of ORR, the adsorbed free energies of various intermediates species (∆ * , ∆ * , ∆ * ) on Co-TCNQ monolayer were estimated. The free energy profile of Co-TCNQ monolayer towards OER under acidic environment has been depicted in Figure 2(b). At = 1.23 V (vs RHE), the reaction steps OH * and OOH * are endothermic and are thermodynamically unfavorable. Hence additional potential equal to the overpotential ( ) will be needed to promote these reaction steps. As illustrated in Figure 2(b), the OOH * formation step is maximum uphill, ca. 0.6 V over the equilibrium potential which gives an estimation of = 0.6 towards OER for Co-TCNQ based catalyst.
Therefore, with additional electrode potential of 0.6 V over the equilibrium value, i.e., for = 1.83 V, all reaction steps become downhill to drive the OER process spontaneously. In alkaline medium (pH = 14), on the other hand, the proton donor is H2O and the corresponding equations describing ORR steps will be as follows:  Figure 4. It is seen from the figure that for TM like Sc, Ti, V, Cr and Mn, the overpotential is higher than 1 V and hence is unsuitable to promote any energy efficient OER activity. In general, decreases from left to rightward for first transition metal row (except Cu) in the periodic table following the order, Sc < V < Ti < Cr < Mn < Fe < Co < Ni, and the lowest is obtained for Ni (0.48 V vs RHE) followed by Co (0.6 V vs RHE) substrate. However, because of low limiting potential ( ) of Ni-TCNQ monolayer (0.18 V vs RHE), it is found to be ineffective as ORR catalyst. The comparable performances as bi-functional catalyst has been observed for Co-TCNQ and Fe-TCNQ monolayer with marginal differences in their and values which indicates that these two monolayer might be the best choice for both OER and ORR catalysts. Nevertheless, many studies have been conducted to evaluate the ORR and OER activities of Fe-TCNQ based catalyst [26,27], but, in depth report on Co-TCNQ electrocatalyst, especially as bi-functional catalyst is highly limited. We, therefore, will focus here on the bi-functional catalytic performances of Co-TCNQ monolayer  Co-TCNQ monolayer and in promoting the catalytic activity.

Conclusions
In summary, an energy efficient electrocatalyst based on Co-TCNQ monolayer has been explored for both oxygen reduction reaction and oxygen evolution reaction. Density functional theory computations revealed that amongst various transition metal-TCNQ based monolayer catalyst, Co-TCNQ monolayer has the best potential to serve as bi-functional catalyst towards ORR and OER in acidic or alkaline media. An overpotential of 0.6 V (vs RHE) was calculated for the Co-TCNQ catalyst to promote the OER activity in either environment, whereas, onset potentials of 0.34 V (vs RHE) in acidic media were measured for ORR to proceed. The other TM-TCNQ based catalyst surfaces have been predicted to exhibit inferior performances in terms of limiting potential and overpotential to drive the ORR and OER processes, respectively. The two-dimensional sheet-like surface of Co-TCNQ monolayer along with its semi-metallic nature in electronic properties helps promoting easy electron transport through the surface and the strong hybridization between Co-3d orbital with O-2p orbital facilitates activation of O2 on Co-TCNQ surface for excellent catalytic activity. Bader charge analysis revealed a net transfer of 0.27|e| charge from Co centre to the OO * adsorbate during ORR and 0.22|e| charge from central