Utilization of structural high entropy alloy for CO oxidation to CO2

The nanoengineered high entropy alloy (HEAs) catalysts have attracted the attention of the scientific community due to their exceptional characteristics; wide range of compositional tunability and the utilization of low-cost transition metals. During various electrochemical reactions, the oxidation of carbon-mono-oxide (CO) is an intermediate and it acts as a poison to reduce the efficiency of the reactions. A nanocrystalline HEA catalyst (CoFeNiGaZn) is prepared by easily scalable cast-cum-crush method, providing pristine catalyst surfaces. It is capable of catalyzing the CO-oxidation to CO2 with high conversion efficiency (99.8%). DFT calculations show that the high activity of the HEA can be attributed to the presence of a considerable amount of filled states of dxz and dyz orbital near the Fermi level for Ni atoms over the surface. Due to the favourable transfer of electrons from this orbital to the LUMO of reactant molecules, the endothermicity of the rate-determining step is 1.13 eV.


2
The engineered nano or atomic scale catalysts have revolutionized the domain of catalysis by providing high selectivity and activity with long term stability, raising hope for sustainability for industrials applications [1,2]. However, noble metal catalysts, commonly utilized are too precious to make them viable for commercial operation. Therefore, nanoengineered catalysts using low-cost elements (transition metal elements), which must be endowed with high stability and efficiency, are the need of the hour. In this regard, different types of materials are under active consideration as well as investigation, revolutionizing the catalytic science to meet the novel requirements, including graphene for HER reaction [3], electronic waste as a catalyst for water splitting [4], carbon dots for ORR and OER reactions [5] etc. However, limitations exist, such as the yield of the catalyst preparation, lack of efficiency, and lower selectivity, which requires the design and development of new catalysts A new concept called entropy stabilized multicomponent alloys, popularly known as high entropy alloys has emerged and reported to be successful in design of novel catalysts for variety of reactions [6]. The five or more elements are mixed in equiatomic or near equiatomic ratio to form single-phase alloys, which has recently been used to design novel catalysts [7,8]. They have shown tremendous potential due to their vast compositional space comprising almost all the metallic elements in the periodic table, allowing us to tune the chemistry and surface microstructure and provide the unlimited scope of design new multipurpose catalysts. Among the low-cost metals available in the periodic table, the alloys made of 3d-transition metals (TMs) are considered to be the best candidates for the nanoengineered HEA catalysts [6]. The four-core effects of HEAs impart exceptional properties; high entropy providing the single-phase stabilization, sluggish diffusion imparting long-term stability, cocktail effect altering their d-band centre with respect to Fermi energy level, and the lattice distortion modifying energy levels of 3 free electrons [9][10][11]. Therefore, the high entropy alloys (HEAs) are expected to provide an active catalyst because a single-compositional alloy can exhibit multiple reactions due to multielemental bonds dangling over the surface with a unique surface electronic structure favourable for the adsorption [12]. This allows the design gamut of the chemistry of high entropy catalysts with different elements for their electrochemical responses. This includes degradation of azo dyes using AlCoCrTiZn HEA catalyst [13], converting CO 2 into highly reduced hydrocarbon (CH 4 , CH 3 OH) using AgAuPdPtCu HEA catalyst [14], formic acid and methanol to hydrogen fuel [15]. The NH 3 decomposition using high entropy alloy (CoMoFeNiCu) was 20 times higher than the pure Ru catalyst [16]. CO oxidation reaction, one of the important yet simple exhaust treatment reactions, has not extensively been studied on the high entropy catalysts. Preferential CO oxidation is important to purify hydrogen gas where CO concentration of lower than 5 ppm is required [17]. Supported Pt catalyst is the most widely used one for this reaction although the cobalt-based catalysts exist, providing ultra-low temperature activity [18,19]. Among various materials used for this reaction are supported metal oxides, perovskite, spinels, substituted metal oxides etc [20][21][22][23]. Among the emerging class of materials, HEAs can possibly outperform previously known catalysts for CO oxidation reaction. However, the magic chemistry is still unknown as a low volume of work exists for CO oxidation reaction on HEAs [24]. In fact, a basic kinetic investigation is not reported so far, which can, in principle lead to a fair comparison of HEAs with the conventional catalysts. In this work, we have explored the CO oxidation activity on low cost easily scalable CoFeZnNiGa HEA nanoparticles. The kinetic parameters have been estimated and compared with the other known catalysts in the literature. The nanocrystalline CoFeZnNiGa HEAs has been prepared by the method of cast-and-crushing, involving cryomilling of the cast alloy ingot at extremely low temperature [25,26]. Therefore, 4 the surface of the catalyst remains native or uncapped due to low temperature milling, making it viable for catalytic applications [25,27]. Further DFT calculations were done to find the active site over the HEA surface along with the probable mechanism, where mainly the effect of suitable orbital overlap among the HEA and reactant molecule was considered.   The HEA nanoparticles are further investigated for heterogeneous gas-solid reaction in the unsupported state. The CO-oxidation reaction is carried out under stoichiometry conditions, the details of gas mixture containing 10% carbon monoxide, 10% Oxygen and remaining Nitrogen.

Results and discussion
The conversion profile of CO over CoFeZnNiGa as a function of the temperature is shown in Figure 2(a). All profiles are recorded in the temperature range from room temperature to 400 °C and using the CO/O 2 ratio of 2:1. The dependence of CO conversion on the flow rates was studied and the results are shown in Figure 2(b). As the space velocity was increased from10,000 -40,000 h −1 (GHSV), no effect was observed on the CO conversion profile and temperature. It concludes that external diffusion (mass transfer) has a limited role on the catalytic activity. This is partially due to the fact that HEA nanoparticles do not contain any macroscopic porosity (as its free-standing nature).
Maximum % conversion of CO over CoFeZnNiGa was found out to be 99.8% at 175 0 Cat 10,000 h -1 GHSV. The rate of CO oxidation reaction was also calculated at various temperatures from the slope of the curve (W/F) with the molar conversion of CO (Figure 2c). Here, W is the weight of the catalyst, and F is the flow rate of CO in sccm [19,28,29]. Rates of the reaction w.r.t weight at temperatures 313K, 348K, 373K were found out to be 2.38, 5.26, 6.39 µmol/g/sec, respectively. Activation energy was found to be 3.761 kJ/mol, which is the lowest among the materials well-known in literature (Figure 2d). Cleary, the HEA catalyst is highly active for the CO oxidation reaction.
To shed light on the probable mechanism and active sites for reaction, density functional theory (DFT) calculations were performed. In the first step, FCC bulk structure of Ni was taken from Materials Project [30] to generate the FCC-HEA surface. Optimized lattice parameters are (a=b=c=3.52 Å) which is in good agreement with previous report [31]. The (111) facet was selected for DFT calculations, as it shows the most prominent pick in the XRD pattern. It was generally believed that for CO oxidation (CO + OCO 2 ), O 2 adsorption takes place on the surface. Its dissociation follows this on transition metals as O 2 can easily dissociate on all transition metals at room temperature [32]. Therefore, the reaction follows Langmuir-   (Figure 3(a,b)). E F of metallic CoFeZnNiGa alloy surface is -4.29 eV, which is energetically lower from E LUMO (-0.24 eV) of O 2 than E LUMO (-2.71 eV) of CO. Therefore, electron transfer from the catalytic surface to O 2 is more difficult than to CO. Adsorption of CO on the surface depletes the electron availability for the activation of O 2 . Hence, the adsorption and the dissociation of O 2 were considered to be the first step followed by adsorption of CO. To 9 understand the most favourable adsorption sites of O 2 and CO, their orbital interaction with the surface was considered. For both O 2 and CO, the LUMO is Π* orbital and hence more prone to interact with the p x or p y orbitals of Ga and d xz or d yz orbitals of Co, Ni, Fe, Zn atoms on the surface Figure 3(c,d). Therefore, a particular atom having more filled states contributed from previously considered orbitals near the Fermi level has more potentiality to transfer the electron In the next step, a similar exercise was done to find out the most suitable adsorption site for CO on the surface containing O*. A particular hollow site containing only Ni-atoms next to the adsorption sites of O maintained the aforementioned criteria (Figure 4(b). After the adsorption of CO, in the neighbouring O site on the CoFeZnNiGa alloy surface, the structure was optimized.
The free energy diagram was calculated to understand the rate-determining step in the considered mechanistic pathway ( Figure 5). Adsorption of O on the surface is less endothermic (0.56 eV) as compared to that of (1.13 eV) of adsorption of CO. This is obvious as initially adsorbed O pulls surface electrons, making the stabilization CO difficult. Desorption of CO 2 from the surface is highly exothermic, putting CoFeZnNiGa alloy surface forward as an active catalyst for CO oxidation reaction.

Conclusions
Catalytic CO oxidation over low-cost novel CoFeZnNiGa HEA nanoparticles, synthesized by cast-cum-crush method was carried out in the presence of the feed oxygen. The conversion and selectivity curves of CO oxidation at different GHSV indicate a maximum CO conversion of 99.8% was achieved at 175°C and 10,000 GHSV. Activation energy was found to be 3.8 kJ/mol, which is the lowest among the well-studied catalysts in the literature. Rates of the reaction at temperature 313k, 348k, 373k are found out to be 2.38, 5.26, 6.39 mol/g/sec, respectively. DFT calculations showed the high activity of Ni hollow sites for CO conversion where the rate-   % Conversion (X) of CO was calculated using the following equations [35] : X CO CO in CO out CO in 100 First, CO was allowed to adsorb on samples at a flow rate of 15ml/min for 30 min at room temperature. The catalyst was then flushed with nitrogen for 30 min at room temperature to remove the gas phase and weakly adsorbed CO. Finally, the catalyst was heated from 25 °C to 400 °C at a constant heating rate of 10°C/min in the presence of nitrogen and desorbed CO and CO 2 was monitored. For the quantification of Conversion of CO, O 2 and the reaction product, calibration of GC was done. A calibration experiment was performed in which the standard gas mixture (CO, CO 2 , H 2 , N 2 ) was used. 1 ml of this calibration standard gas was injected in GC, and the areas of different peaks of the gases were noted for the known concentration of the gases.
These values were used to find out the moles of each gas in CO oxidation reaction.

Simulation methodology
Density functional theory (DFT) were done with "Vienna ab initio simulations (VASP)" package [36]. The Electron-ion interactions were described using all-electron projector augmented wave pseudopotentials, and "Perdew-Bruke-Ernzehof (PBE)" "generalized gradient approximation (GGA)" was used to approximate the electronic exchange and correlations [37]. The plane-wave kinetic energy cut off of 520 eV was used. All the structures were optimized using a conjugate gradient scheme until the energies, and the components of forces reached 10-5 eV and 0.01 eV