Unveiling the magnetic ordering effect in La-doped Ti₃C₂O₂ MXenes on electrocatalytic CO₂ reduction

Abstract

In the pursuit of sustainable energy solutions, the carbon dioxide reduction reaction (CO₂RR) holds immense promise for converting CO₂ into valuable chemicals and fuels. In this view, the exploration of magnetic MXene catalysts is crucial for understanding their reactivity and performance in electrochemical reactions to improve the CO₂ conversion process. Herein, two distinct magnetic configurations, ferromagnetic (FM) and antiferromagnetic (AFM), were considered for La-doped Ti₃C₂O₂ (La-Ti₃C₂O₂). Using density functional theory (DFT) calculations, the first principles simulation was carried out to evaluate the electronic properties, magnetic properties, and CO₂RR potential of these configurations. Our findings reveal an enhancement in semiconductivity and surface reactivity of the La-Ti₃C₂O₂ catalyst, resulting in improved electron transfer characteristics. This facilitates CO₂ adsorption and decreases the formation energy barrier of intermediate species towards the CO₂ hydrogenations. The La-Ti₃C₂O₂ catalyst showed a better performance than the parent molecule Ti₃C₂O₂, which suffers from an insufficiency of reactivity on its surface. Furthermore, the study demonstrates the AFM structure of La-Ti₃C₂O₂ to be the soundest, which thereby displays a better efficiency than the FM structure. Considering our findings, during the current CO₂ conversion process, the reaction pathway with a less energy consumption must be preferred over others.

---

1. Introduction

One of the primary roles of greenhouse gases in the atmosphere is to maintain the earth's temperature within a range of −21 °C to 14 °C, thereby preventing the earth from turning into a frigid planet.¹ However, the current rapid surge in greenhouse gases resulting from the excessive utilization of fossil fuels is leading to global warming.^2,3 With its status as one of the most prevalent greenhouse gases, CO₂ at high concentrations in the atmosphere also pose a serious threat to human health.^4–6 Consequently, the conversion of CO₂ into less problematic and beneficial compounds is both obligatory and urgently needed.^7–9 Numerous reports confirm that CO₂ could be significantly mitigated through substantial advancements in electrocatalytic reactions.^10–13 Fan Lei et al. succeeded in converting CO₂ into formic acid (HCOOH) of a high purity and concentration by an electrocatalytic technique.¹⁴ Carbon monoxide (CO), formaldehyde (CH₂O), methanol (CH₃OH), and methane (CH₄) may also be formed during the electrocatalytic CO₂ reduction process.^15–18 However, the effectiveness of CO₂ reduction through electrocatalysis hinges on the presence of suitable and robust catalysts. While metal catalysts, such as Ru,^19–22 Ir,^23,24 Au,²⁵ Ag,^26,27 Pt,²⁸ Cu,^29,30 and TiO₂ (ref. 31 and 32) showed promise in electrocatalysis, they have fallen short of both CO₂ capture and hydrogenation processes despite diligent efforts. The introduction of graphene catalyst provided a significant boost,³³ but further endeavors are necessary as the efficiency of CO₂ reduction has yet to reach a milestone worth celebrating. With the new 2D MXene materials that have properties similar to graphene and also a large reactive surface area,³⁴ metallic semiconductivity^35,36 and hydrophobicity,³⁷ the electrocatalytic reaction is experiencing a better progress.^{27,28,38–40}

MXenes are derivatives from layer compounds of formula M_n+1AX_n called MAX phase compounds, where M is a transition metal, A is mainly an element of group IIIA or IVA, X is carbon or nitrogen, and n = 1, 2, or 3.⁴¹ Mxenes are obtained by etching layer A of MAX, which results in the Mxene general formula of M_n+1X_n. To achieve this, MAX is placed in an etching agent (HF, NH₄HF₂, ZnCl, and Ti₄N₃) for an appropriate time and then washed quite a few times with deionized water.^42–50 Spectroscopic studies have revealed that the synthesis of Mxenes typically results in the formation of functional groups Tx (Tx = O, OH, and F) on their surfaces.^45,51–57 These atom-terminated functional groups are very active^43,46,58 and can contribute to the unique properties of MXenes that found applications in several areas such as energy storage,^59–62 catalysis,^63–66 electronics,^67–69 water purification,^70,71 thermoelectricity^72,73 and so on. Among the atoms-terminated, O seems to be the most stable as others tend to undergo conversion into O under high temperatures.^46,74 The oxygen on the surface of MXenes is an activated site for diverse reactions.^75,76 Studies showed that heteroatom doped carbon-based catalysts have fascinating performances such as low applied voltage,⁷⁷ high Faraday efficiency^78,79 and long cycle life,⁸⁰ and Anmin Liu et al. demonstrated that nitrogen-doped graphene shows high performance in the CO₂RR.³³ The spintronic field that studies the behaviors of electron charge and spin initiated the idea of investigating the magnetic effect of catalysts in catalysis. Neng Li et al. demonstrated that Fe and Ni used together in high spin and low spin, respectively, increase the catalyst performance for nitrogen reduction.⁸¹ The effect of spin in MXene catalysts on electrocatalytic CO₂RR is, therefore, timely to be explored for CO₂ utilization. According to the first-principles calculations, most of the MXenes are nonmagnetic, though some magnetic MXenes have been synthesized, such as V₂C(OH)₂, V₂CF₂, Cr₂C(OH)₂, and Cr₂CF₂, which are AFM.⁸² To find more magnetic MXenes, a new variety with a double transition metal of M_nM'X_n was synthesized. Jianhui Yang et al. investigated the magnetic effect of double transition metal MXenes and found that the compounds Cr₂VC₂Tx are FM.⁸²

Distinguished as benchmark MXenes, Ti₃C₂Tx stand out as the most readily synthesized variant, making them abundant and cost-effective for a wide range of applications. They boast impressive performance in electrochemistry, energy storage and gas sensor applications.^83–85 However, their efficacy in CO₂RR necessitates improved surface reactivity to effectively respond to spontaneous CO₂ capture and hydrogenation reactions.⁸⁶ Additionally, they exhibit minimal magnetic properties and do not fall into the coveted magnetic material category. Based on the findings of Iqbal Mehroz et al., upon lanthanum (La³⁺) doping into the Ti₃C₂Tx system, they display an improvement in electronic properties and undergo a transformation into magnetic materials, capable of displaying both FM and AFM configurations.⁸⁷ Considering the boosted electronic and magnetic properties of La-Ti₃C₂Tx, and the presence of stable O-terminated functional groups, La-Ti₃C₂O₂ is adopted as the catalyst material for conducting this simulation study on CO₂RR. The objective is to master the challenging processes of spontaneous CO₂ adsorption and hydrogenations, leading to the generation of valuable compounds and hydrocarbons. In this regard, we first assess the CO₂ reduction performance using the magnetic La-Ti₃C₂O₂ in comparison to its nonmagnetic counterpart Ti₃C₂O₂. Subsequently, we conduct an efficiency comparison study between the FM and AFM structures of La-Ti₃C₂O₂. Identifying an optimal pathway is crucial for enhancing the efficiency of electrocatalytic CO₂RR, as certain pathways may offer energy savings compared to others. Hence, we investigate the probable routes taken by the reduction reactions, aiming to select the pathway characterized by lower energy consumption for the CO₂ reduction process.

2. Modeling and simulation methods

Extensive studies conducted on Ti₃C₂ by Shein et al. revealed an FM alignment in the top Ti1 atom sheet that antiferromagnetically paired with the alignment in the bottom Ti1 sheet, while the internal Ti2 atom sheet was nonmagnetic. The ground state of Ti₃C₂ is, therefore, AFM and the atomic magnetic moment is 0.74μ_B per Ti atom.⁸⁸ The O-functional group attached to the surface of Ti₃C₂ creates a weak FM effect⁸⁷ and in this work, we consider Ti₃C₂O₂ as a nonmagnetic material (Fig. 1a). The doping of Ti₃C₂O₂ with La atoms at the Ti sites engenders disorder in electron distribution since La³⁺ is larger than Ti⁴⁺. This increases the saturation magnetization and enhances the magnetic effect of La-Ti₃C₂O₂.⁸⁷ Using the first principles calculations resulting from the DFT implemented in Materials Studio Castep, we perform this simulation study with two La atoms in a Ti₃C₂O₂ system using a 3 × 3 × 1 supercell model, as shown in Fig. 1b. Note that La-Ti₃C₂O₂ is magnetic and either a FM or AFM structure could be observed via the adsorbed La atoms, as shown in Fig. 1c and d. The calculated ground state energy showed stable AFM and FM structures with a difference in energy of 4.4 meV. The modelling vacuum thickness was about 15 Å following the z-direction, the cut-off energy was set to 450 eV, and the k-point was set to 3 × 3 × 1. The Perdew–Burke–Ernzerhof (PBE) functional was used regardless of the exchange and correlation energy. The Gibbs free energy of reaction (ΔG) was calculated using the formulae below:

ΔG = ΔE_DFT + ΔZPE − TΔS (1)

where ΔZPE is the zero-point energy, TΔS is the vibrational entropy at a given temperature and ΔE_DFT is the reaction energy given by the DFT calculations.

Fig. 1 The view of Ti₃C₂O₂ (a) and La-Ti₃C₂O₂ (b) and the corresponding FM (c) and AFM (d) structures of La-Ti₃C₂O₂.

E _DFT (MXene…S) = total energy of the MXene and adsorbed species.

E _DFT (MXene…S–H) = total energy of the MXene and adsorbed species binding with H.

In particular, the binding energy ΔE_DFT for CO₂ and H* is expressed as follows:

ΔE_DFT (CO₂) = E_DFT (MXene….O₂) − E_DFT (MXene) − E_DFT (CO₂) (3)

3. Electronic properties of materials

3.1 Contribution of La atoms in the Ti₃C₂O₂ system

The geometry optimization of La-Ti₃C₂O₂ in Fig. 2a and b confirms that the doping of Ti₃C₂O₂ with the two La atoms engenders disturbance and defect in electron distribution. Fig. 2a represents the structure of Ti₃C₂O₂ where we can see a perfect distribution of all the atoms. The two Ti atoms to be substituted (Ti1 and Ti2) form right angles with the Ti closers, and the oxygen atoms at the surface are well distributed. The presence of La1 and La2 at Ti1 and Ti2 sites disrupts the distribution of the atoms in the material. La1 and La2 endure displacement and move above from their initial positions, as shown in Fig. 2b. The surface oxygen atoms of the La-Ti₃C₂O₂ material undergo disorganization, and the oxygen (Ov) on top, center and closer to La1 and La2 could serve as an activated site for the adsorption of the CO₂.

Fig. 2 The geometry optimized structure of Ti₃C₂O₂ (a) and La-Ti₃C₂O₂ (b), the density of states of d-orbital of La and p-orbital of Ov (c), and the density of states of d-orbital of the 2 substituted Ti and p-orbital of Ov (d).

La-Ti₃C₂O₂ crystallized in the hexagonal structure system,⁸⁷ and the density of states (DOS) diagrams in Fig. 2c and d support the idea of an overlap between the d-orbitals of La atoms and the p-orbital of Ov around the Fermi level, while no overlap phenomena are observed between Ov and the two-substituted Ti in Ti₃C₂O₂. This is because the contribution of La atoms has more impact on the material surface, which thereby displays evidence that Ti₃C₂O₂ shows an enhancement in electronic properties when it is doped with La atoms. Additionally, the peak of Ov located around −2.6 eV in Fig. 2d is closer to the Fermi level compared to the peak around −4.5 eV observed in Fig. 2c. This translates that Ov is more activated in La-Ti₃C₂O₂, and could facilitate the CO₂ binding process.

3.2 Electronic and magnetic properties of the La-Ti₃C₂O₂ material

The electron density difference of La-Ti₃C₂O₂ (both FM and AFM) in Fig. 3a is set considering the yellow color as electron density accumulation and the blue color as electron density loss. It shows that electron density is enriched around Ov and La1 at the edge, while the electron density is depleted around the inner La2. This is confirmed by the Bader charge analysis in Fig. 3b, which clearly depicts the negative charge on La2, indicating that La2 has transferred electrons to the surface oxygen atoms (especially Ov). The electron enrichment on Ov and La1 is harnessed to efficiently capture and hydrogenate CO₂. On the other hand, this electron enrichment is a key aspect of designing efficient catalysts for CO₂RR and achieving the desired product selectivity. Fig. 3c reveals through Bader charge analysis that in Ti₃C₂O₂, Ti1 and Ti2 do enhance the electron density around Ov. Yet, their contributions may not be sufficient to induce the spontaneous capture and hydrogenation of CO₂, especially when compared with the La-Ti₃C₂O₂ case. In the La-Ti₃C₂O₂ configuration, both La1 and Ov are capable of donating electrons to CO₂, which appears to be a more conducive environment for these reactions. The band structure study of La-Ti₃C₂O₂ in Fig. 3d provides 0.227 eV and 0.220 eV as the band gaps of FM and AFM structures, respectively. La-Ti₃C₂O₂ exhibits a short distance between the valence band and the conduction band, indicating its strong semiconductor characteristics in both FM and AFM forms.

Fig. 3 The surface electron density difference of La-Ti₃C₂O₂ (a); the Bader charge of La1, La2, and Ov in La-Ti₃C₂O₂ (b); the Bader charge of Ti1, Ti2 and Ov in Ti₃C₂O₂ (c); the band structure (d), the total density of states (e) and the partial density of states of La (f) of La-Ti₃C₂O₂.

Fig. 3e and f provide indications about the density of states of spin-up (alpha) and spin-down (beta) in La-Ti₃C₂O₂. Fig. 3e is the total density of states and Fig. 3f is the partial density of states of La atoms. It can be seen that the peaks of the spin-up and spin-down have the same appearances and intensities in both figures. This is because the electronic structure of La-Ti₃C₂O₂ allows electrons with both spin-up and spin-down orientations to occupy the same energy levels or states with equal probability. In that case, the spin alignment that corresponds to this material is AFM. This supports the slight difference in ground state energy and band gap of La-Ti₃C₂O₂ (AFM), which could heighten the efficiency of the catalyst in the electrocatalytic CO₂RR.

4. Results and discussion

4.1 Electrocatalytic CO₂RR process

Fig. 2 and 3 illustrate La-Ti₃C₂O₂ as a promising candidate for CO₂ capture and subsequent hydrogenation reactions. Depending on the number of H⁺/e⁻ pairs supplied by the catalyst, we obtain different products according to the considered reaction pathways. The CO₂ reduction initiates with a physisorption step, where CO₂ is adsorbed onto the activated catalyst material without altering its molecular structure. This is followed by a chemisorption reaction during which the CO₂ structure is altered, and the catalyst donates the first H⁺/e⁻ pair to CO₂. The resulting species can be either OCHO˙ (i) or COOH˙ (ii). In path (ii) the second hydrogenation results in the formation of HCOOH (v), or CO (vi) is obtained after the desorption of H₂O. Meanwhile, the attachment of the second proton on OCHO˙ (in i case) offers two possible outcomes. (iii) The proton can be placed on the carbon (˙OCH₂O˙) or (iv) the proton binds the oxygen (HCOOH). In all the previously considered pathways, formaldehyde is obtained at the fourth hydrogenation. Another proton approaches CH₂O and the binding process is either CH₂OH (vii) or CH₃O˙ (viii). The hydrogenation of CH₃O˙ (stable) occurs following two possibilities: (ix) hydrogenation with the proton binding the oxygen (CH₃OH) or (x) the proton approaches and binds the carbon of CH₃O˙ followed by methane (CH₄) release and O remains on the catalyst. From (ix), the hydrogenation of methanol leads to the formation of CH₃˙ and H₂O. The radical CH₃˙, combined with H⁺/e⁻, turns to CH₄. In all the cases, the catalyst material is regenerated after the desorption of CH₄ or H₂O, as shown in Scheme 1.

Scheme 1 Reaction pathways used in the CO₂ reduction process.

4.2 Performance of the magnetic La-Ti₃C₂O₂ MXene in CO₂RR

4.2.1 Approach of the CO₂ binding process.. Once the CO₂ reduction pathways are well defined, the performance of the involved catalyst materials during the process is discussed. First, the efficiency comparison study is carried out between La-Ti₃C₂O₂ and Ti₃C₂O₂, and later the efficiency comparison between the FM and AFM structures of La-Ti₃C₂O₂. The main challenges in CO₂RR arise from the spontaneous capture and the initial hydrogenation of the CO₂ molecule by the material. The capture process at first occurs by the attachment of CO₂ on the material surface through O_V without disturbance of the substrate. The CO₂ adsorption possibilities by O–O_V or C–O_V are investigated to determine the better approach to the reaction. In this regard, the FM and AFM structures of La-Ti₃C₂O₂, and Ti₃C₂O₂ are used to perform the two-adsorption possibilities, and the results are displayed in Fig. 4. It shows that with the three materials, the binding by the oxygen of CO₂ onto Ov (O–O_V) necessitates higher energy of reaction than the binding by the carbon of CO₂ onto Ov (C–O_V). This could be explained by the fact that the bond C–O_V is polarized towards oxygen, while the bond O–O_V has no polarity. The attraction exerted by Ov on the bond electron pair C → O_V promotes the spontaneous capture of the CO₂ molecule on the surface of the material. Additionally, carbon atoms often have a stronger chemical affinity for CO₂ compared to oxygen. This makes the carbon site more favorable for CO₂ binding. A comparison study shows that the required adsorption energy is lower with La-Ti₃C₂O₂ (both FM and AFM) compared to Ti₃C₂O₂, and the energy with the AFM structure is the lowest. This is due to the enhanced surface reactivity of La-Ti₃C₂O₂, and the steadiest AFM configuration, shown in Fig. 2 and 3.

Fig. 4 Physisorption of CO₂ on materials.

After selecting the carbon site of CO₂ for the physisorption process, a chemical reaction takes place thereafter with this time deformation of CO₂. The calculated chemisorption energies of −1.75 eV and −1.87 eV for the FM and AFM configurations, respectively, describe the feasibility of the binding process and confirm that La-Ti₃C₂O₂ is active to accomplish the CO₂ capture reaction. Fig. 5a is the density of states of the free CO₂ molecule, where we can see a peak on the Fermi level. This peak suggests that there is a high density of electronic states related to the molecular orbitals, which may be associated with the π and σ bonding of CO₂. This peak is specific to the electronic structure and potential reactivity of CO₂ but does not imply metallic conductivity, as seen in solid materials. Fig. 5b–d show a deformation in the adsorbed CO₂ structure, and confirm the chemical reactions that occurred between CO₂ and the materials. The ability of CO₂ to react with the materials depends on the involved material. In Fig. 5e we can see that the peak (−1.55 eV) of p-orbitals of CO₂ adsorbed onto the AFM and FM configurations of La-Ti₃C₂O₂ is closer to the Fermi level than the peak (−3.05 eV) of CO₂ adsorbed onto Ti₃C₂O₂, which is 2 times as far away from the Fermi level. As a result, CO₂ becomes more activated, and its reactivity with La-Ti₃C₂O₂ is significantly enhanced. Fig. 5f shows the curves of the density of states of the bond C–O (in CO₂) and the bond C–Ov (bond between Ov and C of CO₂) in both FM and AFM structures of La-Ti₃C₂O₂. Overall, the appearance and specific peaks of C–Ov match with the normal bond of C–O. The large peaks within a range of −5 to −1.05 eV of C–Ov could be explained by a significant concentration of available energy states for electrons provided by the catalyst. This approves that the bond C–Ov is truly a normal C–O bond, and confirms that CO₂ is effectively linked onto La-Ti₃C₂O₂ through Ov. In addition, the substantial concentration of electrons observed through this large peak could act as a facilitator to initiate the hydrogenation processes.

Fig. 5 The DOS of the free CO₂ molecule (a), the DOS of CO₂ adsorbed on Ti₃C₂O₂ (b), La-Ti₃C₂O₂ (AFM), (c) and La-Ti₃C₂O₂ (FM) (d), the DOS comparison (e) and the appearances of the C–O bond in CO₂ and the bond between C of CO₂ and Ov onto La-Ti₃C₂O₂ (f).

4.2.2 CO₂ hydrogenation reactions. The first hydrogenation of CO₂ presents a significant challenge and is considered pivotal to the electrocatalytic CO₂RR process, as it serves as the foundation for subsequent hydrogenation processes. This first hydrogenation begins with the interaction of the CO₂ molecule with the catalyst material, leading to the transfer of the H⁺/e⁻ pair from the material to CO₂. Fig. 6 shows the transition process from CO₂ to the production of OCHO˙, and the next hydrogenation reactions. The COOH˙ formation direction is ruled out in this research because it requires excessive energy, and the entire CO₂RR process for obtaining CH₄ is set considering directions “i, iii, viii, x” which will be elucidated later. Fig. 6a and b demonstrate that the CO₂ adsorption on Ti₃C₂O₂ exhibits structural stability with no significant changes in the CO₂ molecule itself. The electron density difference and the Bader charge analysis emphasize the minimal changes in atom charge distribution (compared with Fig. 3c). This can be attributed to the weak interaction between Ti₃C₂O₂ and CO₂, indicating a lack of reactivity on the material's surface for CO₂ adsorption. Notably, the carbon electrons of CO₂ show a preferential transfer toward the oxygen atoms of CO₂ due to the difference in electronegativity. The electron density difference of La-Ti₃C₂O₂ (AFM and FM) and the related Bader charge analysis in Fig. 6c and d, respectively, reveal a notable alteration of the CO₂ structure, indicating a strong interaction between the materials and CO₂ (compared with Fig. 3a and b). Presently, La1, having gained electrons, exhibits an attraction towards the oxygen in CO₂, prompting a displacement above its optimized position. Consequently, La1 and La2 relinquish electrons in favor of oxygen atoms of CO₂, while Ov, in turn, engages in interaction with the carbon component of CO₂. Examining the electron transfer in AFM and FM scenarios reveals that in the AFM case, the carbon in CO₂ undergoes greater electron loss in favor of Ov. This can be attributed to the heightened attraction exerted by Ov on the carbon, promoting CO₂ adsorption. Additionally, oxygen atoms in CO₂ gain more electrons in the AFM case, potentially playing a crucial role in the subsequent first hydrogenation process.

Fig. 6 The surface electron density difference (a) and the Bader charge analysis (b) of CO₂ adsorbed on Ti₃C₂O₂; the surface electron density difference (c) and the Bader charge analysis (d) of CO₂ adsorbed on La-Ti₃C₂O₂; the chemisorption and first hydrogenation (e) and reaction pathway for CO₂RR (f).

This first hydrogenation is promoted through the low energy barrier of the reaction between the catalyst and CO₂. The calculated Gibbs free energy to obtain OCHO˙ gives −1.72 eV and −1.94 eV for La-Ti₃C₂O₂ (FM) and La-Ti₃C₂O₂ (AFM), respectively, against the 2.62 eV for Ti₃C₂O₂ evinced in Fig. 6e. It can, therefore, be seen that the needed energy for OCHO˙ formation becomes lower when using the La-Ti₃C₂O₂ catalyst, while the reaction with Ti₃C₂O₂ struggles to trigger due to the high reaction energy barrier. This corroborates that the incorporation of La atoms into the Ti₃C₂O₂ structure represents a groundbreaking solution to the difficult hurdle associated with the initial hydrogenation of CO₂. The presence of La enhances the material surface and then facilitates the electron supply from the catalyst to the substrate for the reduction process. Given that La³⁺ (0.106 nm) has a larger atomic size than Ti⁴⁺ (0.058 nm), La-Ti₃C₂O₂ boasts a greater surface area, which translates into enhanced adsorption and hydrogenation efficiency. Additionally, the magnetic effect of La-Ti₃C₂O₂ introduces additional electrostatic interactions with CO₂, enhancing electron transfer characteristics. The spins stimulate a targeted orientation of the La-Ti₃C₂O₂ catalyst electrons towards the substrate. Magnetic catalysts are desirable in improving the material's capability to effectively react with the substrate during this CO₂RR. The La-Ti₃C₂O₂ catalyst, therefore, promotes electrocatalytic CO₂RR to produce valuable compounds and hydrocarbons, and shows higher efficiency than Ti₃C₂O₂ (Fig. 6f). Furthermore, we observe that the AFM configuration displays a lower energy barrier of reaction than the FM one and could tame better the difficult action of obtaining OCHO˙. Since the electron spins of the magnetic La-Ti₃C₂O₂ are not polarized in one direction, the AFM alignment is the proper and steadiest state of La-Ti₃C₂O₂. This explains the good proficiency of La-Ti₃C₂O₂ (AFM) in this electrocatalytic CO₂RR investigation.

4.2.3 CO₂ selectivity evaluation. Analyzing the CO₂ selectivity during the electrocatalytic CO₂RR reveals intriguing insights into a material's capability and reactivity, whether it prioritizes CO₂ adsorption or leads to undesired reactions. The central parameter for assessing CO₂ selectivity is the adsorption energy, which gauges the energy required for the adsorption of CO₂ and H* on the material's surface. A lower adsorption energy for CO₂ compared to H* signifies a higher likelihood for CO₂ adsorption and subsequent conversion, making the reaction with CO₂ a more selective process. Herein, we evaluate the performance of Ti₃C₂O₂ and La-Ti₃C₂O₂ (FM and AFM) in CO₂ selectivity through adsorption energy calculations that are displayed in Fig. 7. Our findings indicate that in the case of Ti₃C₂O₂, the material nearly demonstrates a balanced adsorption potential for both CO₂ and H* reactants. This might result in competitive adsorption, and the CO₂RR selectivity can be compromised. However, the relatively lower CO₂ adsorption energy suggests that the material holds promise for CO₂ conversion. Comparatively, it is clearly seen that the adsorption energy for CO₂ is much lower than the adsorption energy of H* when using La-Ti₃C₂O₂. This lower adsorption energy for CO₂ indicates higher selectivity, making La-Ti₃C₂O₂ a more favorable catalyst for CO₂RR. It is essential to highlight that the adsorption energy for H* is marginally higher in the AFM configuration compared to the FM configuration. This distinction suggests that La-Ti₃C₂O₂ (AFM) exhibits greater selectivity for CO₂ conversion. These findings underscore the significance of material design, particularly the introduction of magnetic elements, in tuning the CO₂RR selectivity and advancing the development of efficient catalysts for CO₂ conversion processes.

Fig. 7 The adsorption energy for CO₂ and H* using Ti₃C₂O₂, and AFM and FM configurations of La-Ti₃C₂O₂.

4.3 Favorable pathway of the reduction process

The pathways outlined in Scheme 1 represent potential electrocatalytic directions for CO₂RR during the process. Nevertheless, certain pathways exhibit thermodynamic preference, making them more favorable. Species characterized by lower energy barriers of formation (with all three materials for accurate evaluation) are stable and preferentially obtained (Table 1). The first hydrogenation, where a proton binds to the carbon of CO₂ (OCHO˙), exhibits a significantly lower energy barrier and stronger stability compared to proton placement on the oxygen atom (COOH˙). Consequently, OCHO˙ demonstrates a higher likelihood of formation, making pathway (i) the thermodynamically preferred route. In the second hydrogenation along the path (i), there are two potential outcomes, ˙OCH₂O˙ (iii) or HCOOH (iv). Energy calculations indicate that the formation of ˙OCH₂O˙ is more favorable, as it has a lower energy barrier. Therefore, ˙OCH₂O˙ is selected to proceed with subsequent reactions, ultimately leading to the production of formaldehyde during the fourth hydrogenation. For the fifth hydrogenation, the likely products are CH₂OH (vii) and CH₃O˙ (viii). The radical CH₃O˙ displays a lower energy barrier of formation than CH₂OH, providing further support for the preference of path (viii). Regarding the sixth hydrogenation, it is evident that continuing the reaction with O (lower energy of formation) is more advantageous than with CH₃OH. The path (x) is considered due to its lower energy consumption. Considering the above results, the favorable possibility in this simulation to well approach the CO₂RR is to follow (i), (iii), (viii) and (x) directions. The optimal pathway direction for conducting this electrocatalytic CO₂RR with the enhanced catalyst material La-Ti₃C₂O₂ (AFM) is therefore proposed in Fig. 8.

Table 1 Gibbs free energy of the reaction (with ZPE corrections)

Compounds	La-Ti3C2O2 (FM)	La-Ti3C2O2 (AFM)	Ti3C2O2
CO2	−1.75	−1.87	−0.005
OCHO˙	−1.72	−1.94	2.62
COOH˙	2.39	2.39	0.37
˙OCH2O˙	1.01	1.18	−1.92
HCOOH	2.42	2.60	−1.39
CO	−1.18	−1.18	0.40
HOCH2O˙	0.48	0.36	0.30
CHO˙	0.74	0.71	−0.04
CH2O	1.92	1.90	1.75
CH3O˙	−0.78	−0.62	2.30
CH2OH	1.03	1.20	−0.46
O	−0.10	−0.11	0.06
CH3OH	1.14	1.08	−1.46
OH˙	−1.30	−1.26	−1.47
H2O	1.60	1.58	−2.15
CH3˙	2.04	2.12	2.44
CH4	−0.78	−0.80	−1.77

---

Fig. 8 A favorable reaction pathway for CO₂RR with La-Ti₃C₂O₂ (AFM).

5. Conclusion

First principles calculations using DFT were employed to assess the effectiveness of electrocatalytic CO₂RR using MXene catalysts. Simulations carried out with La-Ti₃C₂O₂ affirmed the materials' capability for spontaneous CO₂ capture and hydrogenation reactions. La-Ti₃C₂O₂ exhibits an enhancement in electronic properties and surface reactivity, which contributes to reducing the formation energy barrier of intermediate species during electrocatalytic CO₂RR. The magnetic effect of La-Ti₃C₂O₂ enhances the material's ability to react with the adsorbed substrate and assist the CO₂RR, resulting in the production of hydrocarbons and other valuable compounds. This advancement addresses both ecological and energy-related concerns. Furthermore, the investigations have unveiled that the tangible magnetic structure of La-Ti₃C₂O₂ is AFM and thereby exhibits superior performance than the FM configuration. Additionally, it is essential to carry out the electrocatalytic CO₂ reduction while considering the direction that consumes less energy, as this aspect is of utmost usefulness for efficiency.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work is supported by the Natural Science Fund for Distinguished Young Scholars of Hubei Province (No. 2020CFA087); Guangdong Basic and Applied Basic Research Foundation (No. 2022A1515011303), the Basic Research Program of Shenzhen (No. JCYJ20190809120015163), the Central Government Guides Local Science and Technology Development Funds to Freely Explore Basic Research Projects (2021Szvup106), and the Fundamental Research Funds for the Central Universities.

References

1. T. R. Anderson, E. Hawkins and P. D. Jones, Endeavour, 2016, 40(3), 178–187.
2. M. Meinshausen and W. Hare, et al. , Nature, 2009, 458, 1158–1162.
3. W. Peng and W. Zhou, et al. , Sci. Total Environ., 2018, 619, 1163–1169.
4. C. Duarte, J. Łukasz and J. Mariusz, Front. Public Health, 2020, 8, 543322.
5. A. Jacobson Tyler, J. S. Kler, T. Hernke Michael, K. Braun Rudolf, C. Meyer Keith and W. E. Funk, Nat Sustain., 2019, 2, 691–701.
6. N. Hudda and S. A. Fruin, Sci. Total Environ., 2018, 610, 1448–1456.
7. U. Terranova, F. Viñes, N. H. De Leeuw and F. Illas, J. Mater. Chem. A, 2020, 8, 9392–9398.
8. E. Trenberth Kevin, T. Fasullo John and K. Jeffrey, Bull. Am. Meteorol. Soc., 2009, 90(3), 311–323.
9. R. B. Jackson, P. Friedlingstein, R. M. Andrew, J. G. Canadell, C. Le Quéré and G. P. Peters, Environ. Res. Lett., 2019, 14, 121001.
10. Y. Wang, B. Xia, H. Qin, F. Wang, Y. Li, X. Li, S. Kang, Y. Zuo and L. Cui, ACS Appl. Mater. Interfaces, 2016, 8(27), 17212–17219.
11. X. Li, J. Yu, J. Mietek and X. Chen, Chem. Rev., 2019, 119(6), 3962–4179.
12. C. Hiragond, H. Kim, J. Lee, S. Sorcar, C. Erkey and S. I. In, Catalysts, 2020, 10(1), 98.
13. N. Li, X. Chen, W.-J. Ong, D. R. MacFarlane, X. Zhao, A. K. Cheetham and C. Sun, ACS Nano, 2017, 11(11), 10825–10833.
14. L. Fan, C. Xia, P. Zhu, Y. Lu and H. Wang, Nat. Commun., 2020, 11(1), 3633.
15. J. Gu, C.-S. Hsu, L. Bai, H. M. Chen and X. Hu, Science, 2019, 364(6445), 1091–1094.
16. K. Nakata, T. Ozaki, T. Chiaki, A. Fujishima and E. Yasuaki, Angew. Chem., Int. Ed., 2014, 53(3), 871–874.
17. F. Sha, Z. Han, S. Tang, J. Wang and C. Li, ChemSusChem, 2020, 13(23), 6160–6181.
18. H. Arandiyan, K. Kani, Y. Wang, B. Jiang, J. Kim, M. Yoshino, M. Rezaei, A. E. Rowan, H. Dai and Y. Yamauchi, ACS Appl. Mater. Interfaces, 2018, 10(30), 24963–24968.
19. S. Wang, E. T. Schrunk, H. Mahajan and R. J. Farrauto, Catalysts, 2017, 7(12), 88_.
20. B. Arghya, H. H. Anton and V. Tejs, J. Phys. Chem. C, 2017, 121(34), 18333–18343.
21. K. Mohammadreza, A. Hansen Heine, R. Jan and K. Nørskov Jens, ACS Catal., 2015, 5(7), 4075–4081.
22. A. Jaleel, S. H. Kim, P. Natarajan, G. H. Gunasekar, K. Park, S. Yoon and K. D. Jung, J. CO₂ Util., 2020, 35, 245–255.
23. H. Zhao, H. Zhu, Y. Feng, Q. Zhao, B. Suo and W. Zou, et al. , ChemElectroChem, 2020, 7(24), 5036–5043.
24. K. Tang, Z. Wang, W. Zou, H. Guo, Y. Wu, Y. Pu, Q. Tong, H. Wan, X. Gu, L. Dong, J. Rong and Y. W. Chen, ACS Appl. Mater. Interfaces, 2021, 13(5), 6219–6228.
25. S. Zhao, Z. Tang, S. Guo, M. Han, C. Zhu, Y. Zhou, B. Liang, J. Gao, H. Huang, Y. Li, Y. Liu and Z. Kang, ACS Catal., 2017, 8(1), 188–197.
26. J. Rosen, G. S. Hutchings, Q. Lu, S. Rivera, Y. Zhou, D. G. Vlachos and F. Jiao, ACS Catal., 2015, 5(7), 4293–4299.
27. Z. Zhang, H. Li, G. Zou, C. Fernandez, B. Liu, Q. Zhang and Q. Peng, ACS Sustain. Chem. Eng., 2016, 4(12), 6763–6771.
28. X. Xie, S. Chen, W. Ding, Y. Nie and Z. Wei, Chem. Commun., 2013, 49(86), 10112_.
29. G. Yin, M. Nishikawa, Y. Nosaka, N. Srinivasan, D. Atarashi, E. Sakai and M. Miyauchi, ACS Nano, 2015, 9(2), 2111–2119.
30. Z. Changjun, Nat. Energy, 2016, 1(8), 16124.
31. T. P. Nguyen, D. Nguyen, V. H. Nguyen, T. H. Le, D. N. Vo, Q. T. Trinh, S. R. Bae, S. Y. Chae, S. Y. Kim and Q. V. Le, Nanomaterials, 2020, 10(2), 337.
32. M. Tasbihi, K. Kamila, I. Troppová, M. Edelmannová, M. Reli, L. Čapek and R. Schomäcker, Environ. Sci. Pollut. Res. Int., 2018, 25(35), 34903–34911.
33. A. Liu, W. Guan, K. Wu, X. Ren, L. Gao and T. Ma, Appl. Surf. Sci., 2021, 540, 148319_.
34. P. A. Maughan, L. Bouscarrat, V. R. Seymour, S. Shao and S. J. Haigh, et al. , Nanoscale Adv., 2021, 3(11), 3145–3158.
35. B. Lyu, M. Kim, H. Jing, J. Kang, C. Qian, S. Lee and J. H. Cho, ACS Nano, 2019, 13(10), 11392–11400.
36. B. Zazoum, A. Bachri and J. Nayfeh, Materials, 2021, 14(21), 6603.
37. J. Liu, H.-B. Zhang, R. Sun, Y. Liu, Z. Liu, A. Zhou and Z.-Z. Yu, Adv. Mater., 2017, 29(38), 1702367.
38. J. Zhu, H. Enna, G. Zhao, Y. Zhou, D. Huang and G. Yue, et al. , Coord. Chem. Rev., 2017, 352, 306–327.
39. G. Gao, A. P. O'Mullane and A. Du, ACS Catal., 2016, 7(1), 494–500.
40. C. Ling, L. Shi, Y. Ouyang and J. Wang, Chem. Mater., 2016, 28, 9026–9032.
41. R. M. Ronchi, J. T. Arantes and S. F. Santos, Ceram. Int., 2019, 45(15), 18167–18188.
42. M. Naguib, V. N. Mochalin, M. W. Barsoum and Y. Gogotsi, Adv. Mater., 2013, 26(7), 992–1005.
43. M. Naguib, O. Mashtalir, J. Carle, V. Presser, J. Lu, L. Hultman and M. W. Barsoum, ACS Nano, 2012, 6(2), 1322–1331.
44. B. Anasori, Y. Xie, M. Beidaghi, J. Lu, B. C. Hosler, L. Hultman, P. R. Kent, Y. Gogotsi and M. W. Barsoum, ACS Nano, 2015, 9(10), 9507–9516.
45. M. Naguib, M. Kurtoglu, V. Presser, J. Lu, J. Niu, M. Heon and M. W. Barsoum, Adv. Mater., 2011, 23(37), 4248–4253.
46. Y. Xie, M. Naguib, V. N. Mochalin, M. W. Barsoum and Y. Gogotsi, et al. , J. Am. Chem. Soc., 2014, 136(17), 6385–6394.
47. F. Liu, J. Zhou, S. Wang, B. Wang, C. Shen, L. Wang, Q. Hu, Q. Huang and A. Zhou, J. Electrochem. Soc., 2017, 164, A709.
48. J. Halim, M. R. Lukatskaya, K. M. Cook, J. Lu, C. R. Smith, L. A. Näslund, S. J. May, L. Hultman, Y. Gogotsi, P. Eklund and M. W. Barsoum, Chem. Mater., 2014, 26(7), 2374–2381.
49. M. Li, J. Lu, K. Luo, Y. Li, K. Chang and K. Chen, et al. , J. Am. Chem. Soc., 2019, 141(11), 4730–4737.
50. U. Patrick and A. Babak, et al. , Nanoscale, 2016, 8(22), 11385–11391.
51. M. Khazaei, M. Arai, T. Sasaki, C.-Y. Chung, N. S. Venkataramanan, M. Estili and Y. Kawazoe, Adv. Funct. Mater., 2012, 23(17), 2185–2192.
52. I. R. Shein and A. L. Ivanovskii, Micro Nano Lett., 2013, 8(2), 59–62.
53. P. Paul, P. Chakraborty, T. Das, D. Nafday and T. Saha-Dasgupta, Phys. Rev. B, 2017, 96(7), 079903_.
54. G. R. Berdiyorov, Europhys. Lett., 2015, 111(6), 67002.
55. O. Mashtalir, M. Naguib and V. Mochalin, et al. , Nat. Commun., 2013, 4, 1716.
56. J. H. Kristopher, B. Matthieu, M. Naguib, W. Barsoum Michel and R. Goward Gillian, J. Phys. Chem. C, 2015, 119(24), 13713–13720.
57. H. Joseph, M. K. Cook, M. Naguib, P. Eklund, G. Yury, J. Rosen and W. Barsoum Michel, Appl. Surf. Sci., 2016, 362, 406–417.
58. J.-H. YANG, S.-Z. Zhang, J.-L. JI and S.-H. Wei, Acta Phys.-Chim. Sin., 2015, 31(2), 369–376.
59. M. Ghidiu, R. Lukatskaya Maria, M.-Q. Zhao, Y. Gogotsi and M. W. Barsoum, Nature, 2014, 516, 78–81.
60. Q. Hu, H. Wang, Q. Wu, X. Ye, A. Zhou and D. Sun, et al. , Int. J. Hydrogen Energy, 2014, 39(20), 10606–10612.
61. C. E. Ren, M.-Q. Zhao, M. Taron, H. Joseph, B. Muhammad, K. Sankalp, A. Babak, M. W. Barsoum and G. Yury, ChemElectroChem, 2016, 3(5), 689–693.
62. Z. Bao, C. Lu, X. Cao, P. Zhang, L. Yang, H. Zhang, D. Sha, W. He, W. Zhang, L. Pan and Z. Sun, Chin. Chem. Lett., 2021, 32(9), 2648–2658.
63. J. Liu, W. Peng and Y. Li, et al. , Trans. Tianjin Univ., 2020, 26, 149–171.
64. M. Yu, S. Zhou and Z. Wang, et al. , Nano Energy, 2018, 44, 181–190.
65. H. Chen, D. Handoko Albertus, J. Xiao and F. Xiang, et al. , ACS Appl. Mater. Interfaces, 2019, 11(40), 36571–36579.
66. S. Zheng, S. Li, Z. Mei, Z. Hu, M. Chu, J. Liu, X. Chen and F. Pan, J. Phys. Chem. Lett., 2019, 10(22), 6984–6989.
67. H. Kim and H. Alshareef, ACS Mater. Lett., 2020, 2(1), 55–70.
68. N. Li, J. Peng, W.-J. Ong, T. Ma, A. Arramel, P. Zhang, J. Jiang, X. Yuan and C. M. Zhang, Matter, 2021, 4(2), 377–407.
69. Y. Wang, Y. Xu, M. Hu, H. Ling and X. Zhu, Nanophotonics, 2020, 9(7), 1601–1620.
70. X. Xie, C. Chen and N. Zhang, et al. , Nat. Sustain., 2019, 2, 856–862.
71. A. J. Al-Hamadani Yasir, B.-M. Jun, M. Yoon, T.-Q. Nader, A. Snyder Shane, M. Jang, J. Heo and Y. Yoon, Chemosphere, 2020, 254, 126821.
72. M. R. Lukatskaya, O. Mashtalir, C. E. Ren, Y. Dall'Agnese, P. Rozier, P. L. Taberna, M. Naguib, P. Simon, M. W. Barsoum and Y. Gogotsi, Science, 2013, 341(6153), 1502–1505.
73. X. Lu, Q. Zhang, J. Liao, H. Chen, Y. Fan and J. Xing, et al. , Adv. Energy Mater., 2019, 10(2), 1902986.
74. D. Sun, Q. Hu, J. Chen, X. Zhang, L. Wang, Q. Wu and A. Zhou, ACS Appl. Mater. Interfaces, 2016, 8(1), 74–81.
75. L. Liu, H. Zhao, J. M. Andino and Y. Li, ACS Catal., 2012, 2(8), 1817–1828.
76. Y. Ji and Y. Luo, J. Am. Chem. Soc., 2016, 138(49), 15896–15902.
77. B. Kim, A. Franklin, N. Colin, H. Wilfried and S. Tulevski George, Appl. Phys. Lett., 2014, 105(6), 063111.
78. M. Wang, S. Liu and T. Qian, et al. , Nat. Commun., 2019, 10(1), 341.
79. Y. Sun, S. Li, J. Zarko, B. Denis, H. Wang, P. Benjamin, X. wang, S. Kuehl and S. Peter, ChemSusChem, 2018, 11(19), 3388–3395.
80. P. A. Maughan, L. Bouscarrat, V. R. Seymour, S. Shao, S. J. Haigh, R. Dawson, N. Tapia-Ruiz and N. Bimbo, Nanoscale Adv., 2021, 3(11), 3145–3158.
81. N. Li, Z. Wang, P. Zhang, X. Li, A. Arramel, C. Sun, X. Zhou and X. Zhao, J. Mater. Chem. A, 2022, 10(42), 22760–22770.
82. J. Yang, X. Zhou, X. Luo, S. Zhang and L. Chen, Appl. Phys. Lett., 2016, 109(20), 203109_.
83. P. Komen, L. Ngamwongwan, S. Jungthawan, A. Junkaew and S. Suthirakun, ACS Appl. Mater. Interfaces, 2021, 13(48), 57306–57316.
84. M. D. Siriwardane Edirisuriya, D. Ilker, S. Cem and C. Deniz, ACS Appl. Energy Mater., 2019, 2(2), 1251–1258.
85. Y. Yao, Z. Wang, Y. Han and L. Xie, et al. , Sens. Actuators, B, 2023, 381, 133412.
86. F. Li, X. Wang and R. Wang, Catalysts, 2021, 11(2), 161.
87. M. Iqbal, J. Fatheema, Q. Noor, M. Rani, M. Mumtaz, R.-K. Zheng and S. Rizwan, Mater. Today Chem., 2020, 16, 100271.
88. I. R. Shein and A. L. Ivanovskii, Comput. Mater. Sci., 2012, 65, 104–114.

---