Synergistic effect of diatomic materials on efficient formaldehyde sensing and degradation

Abstract

A high-sensitivity sensor for formaldehyde (HCHO) is crucial in environmental detection and human health studies. However, the development of sensing materials with remarkable adsorption capacity remains a challenge. In this study, we employed density functional theory to screen twenty-seven transition metals as potential single atom (SA) and diatomic (DA) adsorption materials for detecting HCHO. Among them, Hf₂-C₂N exhibited excellent HCHO adsorption capability among fifty-four candidate materials due to its more negative adsorption energy (−7.32 eV). Through extensive electronic structure and orbital analyses, we elucidated that the enhanced activity of Hf DAs, acting as the active site for HCHO adsorption, was attributed to the introduction of an additional Hf atom in Hf₂-C₂N. Furthermore, we discussed in detail the degradation process of HCHO on Hf₂-C₂N using transition state analysis which revealed a low potential barrier (2.1 eV), indicating its potential as a high-performance catalyst. This work not only screens a bifunctional catalyst for efficient adsorption and degradation of HCHO but also provides valuable insights for further experimental exploration.

---

1. Introduction

Formaldehyde (HCHO) is a hazardous volatile organic compound (VOC) commonly found indoors, characterized by its colourless and odourless nature, which poses an undetectable threat to human health and acts as a serious carcinogen.^1,2 Moreover, prolonged exposure to HCHO can lead to genetic mutations.³ The World Health Organization (WHO) has set a stringent exposure threshold of 0.1 mg m⁻³ for HCHO,⁴ underscoring the urgency in detecting this compound. Gas sensors assume a pivotal role in the detection of HCHO. However, conventional sensors often suffer from limitations such as poor performance in high humidity environments known as water poisoning.^5,6 Another challenge lies in the selectivity of traditional sensors towards detecting HCHO gas under complex environmental conditions.⁷ Therefore, it is imperative to explore efficient materials for sensing HCHO and catalysts for its degradation.

In recent decades, a range of metal oxide semiconductors (MOSs) have proven to be effective in the detection of formaldehyde (HCHO), such as TiO₂,⁸ In₂O₃,⁹ and SnO₂.¹⁰ However, the low sensitivity of these materials in parts per million (ppm) limits their application in HCHO sensing.¹¹ Two-dimensional materials, owing to their distinct chemical and physical characteristics (layered structure, large surface area, and changeable chemical properties), are considered to be an alternative to conventional sensing materials.¹² Due to their superior surface sensitivity, 2D materials have the most potential for chemical sensing applications, especially in highly sensitive environments.¹³ To further improve the detection sensitivity of HCHO for 2D materials, a lot of interest has been drawn to SAs owing to their enormous specific surface area, high adsorption capacity and potentially low cost.^14,15 Wang et al. prepared V-SAs on ultrathin carbon nitride, which exhibited high activity HCHO oxidation performance (over a span of 180 minutes, the concentration of HCHO decreased from an initial level of 300 ppm to 57 ppm). The experiments and theoretical calculations were combined to explain that the V-SA as the active site promoted the oxidation of HCHO.¹⁶ Shin et al. trapped a Pt SA in a carbon nitrogen/SnO₂ heterostructure, and this 1D nano-heterostructure supported Pt SA system exhibited high sensitivity (response = 33.9 at 5 ppm) and selectivity for the detection of HCHO.¹⁷ In our previous work, Zr-C₃N₆ and Hf-C₃N₆ were identified to exhibit efficient adsorption of HCHO.¹⁸ However, SAs as the HCHO detection material remain short of industrial requirements and the single active site of SAs limit their adsorption for HCHO.

Recently, diatomic (DA) materials have received significant attention, presenting a promising avenue in catalysis beyond SA catalysts. DA materials exhibit increased metal loading and greater adaptability in the configuration of active sites, resulting in enhanced adsorption and catalytic capacity compared to SA catalysts due to synergistic effects between metals.^19,20 The interaction of two adjacent atoms in DA structures effectively stabilizes the intermediates, leading to improved activity and selectivity.^21,22 At present, although the adsorption detection and detection of HCHO by SAs has been demonstrated in experiments^23,24 and simulations,²⁵ DA materials that possibly offer higher performance are still lacking in reports. As a crucial component of DA systems, the substrate plays a vital role. C₂N, a graphene-like material with distributed pores, large specific surface area, and suitable band gap values, is widely employed as a substrate for investigating adsorption and catalytic performances of DA materials.^26–28 However, further research progress has been impeded by the high cost and time requirements associated with standard experimental methods. Therefore, there is an urgent need to employ theoretical simulation methods for screening transition metals anchored on C₂N substrates in both SA and DA configurations for formaldehyde (HCHO) adsorption. More importantly, although there are numerous newly discovered novel materials with highly efficient properties, for practical applications, it is more feasible to prioritize materials that can be easily synthesized rather than those existing solely in theoretical realms.

In this study, we conducted a comprehensive simulation analysis on twenty-seven transition metals anchored on C₂N to investigate their thermodynamic stability and adsorption capacity for HCHO in two different configurations. Among the candidates, Hf₂-C₂N exhibited strong adsorption of HCHO. To understand the underlying mechanism, we analyzed various properties including projected density of states (PDOS), d-band center, electron localization function (ELF), charge density difference (CDD), Bader charge transfer, and crystal orbital Hamiltonian population (COHP). Additionally, the potential degradation ability of Hf₂-C₂N towards HCHO was discussed using transition state theory. This theoretical investigation not only presents a promising material for efficient sensing and degradation of diatomic molecules like HCHO but also provides valuable insights into its adsorption mechanism.

2. Calculation method

All spin-polarized DFT simulations were performed using with the generalized gradient approximation (GGA) method with the Perdew–Burke–Ernzerhof (PBE)²⁹ functional as implemented in the Vienna ab initio simulation package (VASP)³⁰ with van der Waals (vdW) correction (DFT+D3)³¹ based on the MedeA platform. The projector-augmented plane-wave (PAW)³² method was employed to describe the ion–electron interactions and the cutoff energy was set as 450 eV. For the calculation of structural relaxation and electronic information, the K-points were set to 3 × 3 × 1 and 6 × 6 × 1 in the Monkhorst–Pack meshes, respectively. A vacuum layer with a length of 15 Å in the z direction was used to avoid interactions between periodic structures. Up until the convergence thresholds of energy and the Hellmann–Feynman force, which are 10⁻⁵ eV and 0.03 eV Å⁻¹, respectively, structures relaxed. Post-processing of electronic structure information was carried out with the VASPKIT core.³³ The climbing image nudged elastic band (CI-NEB) method was employed to search for the conformation and energy of the transition state in the degradation process of HCHO.³⁴

The binding energy (E_b) of transition metals anchored on pores of C₂N is defined as:

E_b = E_(TMn-C2N) − E_(C2N) − nE_(TM) (1)

where E_(TMn-C2N) is the total energy of transition metals anchored on pores of C₂N, E_(C2N) is the energy of a pristine C₂N surface, and E_(TM) is the energy per atom of TM. The cohesive energy (E_Coh) of transition metals is defined as:

E_Coh = E_(TM-bulk)/N − E_(TM) (2)

where E_(TM-bulk) is the energy of a transition metal unit cell and N is the number of atoms per unit cell. The adsorption energies (E_ads) of the HCHO molecule on TM_n-C₂N are defined as:

E_ads = E_{(TMn-C2N@HCHO)} − E_(TMn-C2N) − E_(HCHO) (3)

where E_{(TMn-C2N@HCHO)} is the total energy of SAs or DAs with HCHO adsorbed and E_(HCHO) is the energy of HCHO.

3. Results and discussion

3.1 Structures of SA-C₂N and DA-C₂N

The C₂N structure was initially relaxed, as shown in Fig. 1a. The homogeneous pore of C₂N is formed by 12C and 6 N atoms into four units. The inter-hole distance for C₂N is 8.300 Å, which agrees with earlier reports.³⁵ Monomolecular C₂N has two different forms of N atoms (pyridine-type and pyrazine-type).³⁶ According to the report by Du et al., the bond lengths of C–C(1) (1.465 Å), C–C(2) (1.428 Å), and C–N (1.335 Å) of C₂N are all in agreement.³⁷ In addition, it is found that monolayer C₂N exhibits electronic properties indicative of a semiconductor, characterized by a direct band gap of 1.67 eV, consistent with prior theoretical investigations.²⁶

Fig. 1 (a–d) The optimised structures of pure C₂N, SA-C₂N and DA-C₂N. (d–f) The binding energies and cohesive energies of third, fourth and fifth period transition metals. (g–i) Band structure and projected density of states of C₂N, Hf-C₂N and Hf₂-C₂N.

Following this, we investigated twenty-seven transition metals anchored on the pores of monolayer C₂N in two forms (SA-C₂N and DA-C₂N), displaying Hf-C₂N and Hf₂-C₂N as representative structures as shown in Fig. 1b and c The transition metal SA prefers to anchor the two nitrogen atoms around the C₂N pore, and the two transition metal atoms of DA incline to anchor the two pairs of nitrogen atoms facing each other in the C₂N pore. The transition metals SA and DA are anchored on the same plane as C₂N. In addition, the pore length of C₂N shortens after single-atom anchoring (from 8.300 Å to 8.134 Å); however, it lengthens when double-atom anchoring occurs (from 8.300 Å to 8.615 Å).

To study the stability of the substrate materials for HCHO adsorption, we investigated the E_b of SAs and DAs anchored on C₂N and the E_Coh of transition metals, and the detailed energies of twenty-seven SAs and DAs are shown in the Tables S2 and S3,† respectively. As shown in Fig. 1d–f, for the twenty-seven transition metals considered, a more negative E_b to E_Coh value indicates more stability for the structure.³⁸ Except for Ag-C₂N, Au-C₂N and Cd-C₂N, E_b values of DA-C₂N are all more negative than their corresponding E_Coh values, demonstrating the thermodynamic stability of DA-C₂N. The E_b values of DA-C₂N are generally more negative than that of SA-C₂N, suggesting that anchoring DAs on C₂N pores is more stable than anchoring SAs (except for Y-C₂N, Ag-C₂N, Cd-C₂N and Au-C₂N). SA-C₂N, particularly the fifth-period transition metals, has more positive E_b values than the corresponding E_Coh values, suggesting that the stability of some SA-C₂N materials is a non-negligible synthetic hindrance. In summary, except for Y-C₂N, Ag-C₂N, Au-C₂N and Cd-C₂N, DA-C₂N materials are stable and universally more stable than SA-C₂N. In addition, the band structure of C₂N shows that the direct band gap of C₂N was 1.666 eV (Fig. 1g), and with the introduction of metal atoms, the band gap decreases sharply, switching from a direct to an indirect band gap (using Hf-C₂N and Hf₂-C₂N as examples, as shown in Fig. 1h and i), which is consistent with previous theoretical and experimental reports.^26,28 The identical band gaps indicate that the structures are efficiently a universal structure for the ensuing adsorption.

3.2 Adsorption characteristics

The adsorption of HCHO on both SA-C₂N and DA-C₂N is of great importance in evaluating their potential as highly sensitive HCHO sensing materials. Subsequently, the adsorption of HCHO on SA-C₂N and DA-C₂N was explored in two distinct adsorption configurations, namely, side-on and end-on orientations. Schematic diagrams of the adsorption configurations are shown in Fig. 2a, where the side-on configuration is HCHO adsorbed on the transition metal site in a parallel way on C₂N and the end-on configuration for vertical direction. According to previous reports, HCHO in the end-on adsorption configuration favoured bound metal atoms with an O atom.³⁹ The results of the adsorption energy of SA-C₂N and DA-C₂N for HCHO are shown in Fig. 2b–d and e–g, respectively. A more negative adsorption energy implies stronger material adsorption for HCHO.^40,41 It was found that the E_ads values of materials are negative except for Mn-C₂N and Mn₂-C₂N, indicating that for these materials the adsorption of HCHO is a thermodynamically spontaneous reaction. Note that for transition metals of the same period, the metals of IIIB and IVB groups anchored on C₂N are more strongly adsorbed to HCHO. In addition, HCHO adsorption on SA-C₂N and DA-C₂N generally favours the side-on configuration owing to the more negative E_ads. For the same transition metal, DA-C₂N generally showed a stronger adsorption of HCHO than SA-C₂N. In these materials above, Hf-C₂N (−3.14 eV) and Hf₂-C₂N (−7.32 eV) showed the most negative E_ads values for HCHO on SA-C₂N and DA-C₂N, respectively. The strong adsorption capacity is suitable for high-sensitivity HCHO sensing materials, and a more negative adsorption energy than −0.5 eV indicates that they adsorb HCHO by chemisorption.^42,43

Fig. 2 (a) The schematic diagram of HCHO absorbed on SA-C₂N and DA-C₂N with different configurations, respectively. (b–d) The adsorption energy of HCHO absorbed on SA-C₂N with side-on and end-on configurations. (e–g) The adsorption energy of HCHO absorbed on DA-C₂N with side-on and end-on configurations. (h–j) The linear relationship of the d-band centre and Bader charge with adsorption energy for different periods of metal anchored DA-C₂N, h, j and k represent 3d, 4d and 5d transition metal anchored C₂N structures, respectively.

Then we investigated the metal d-band centre and Bader charge of HCHO adsorption by DA with different periodic metal anchoring, and the results revealed a linear correlation between the d-band centre and Bader charge concerning adsorption energy, as shown in Fig. 2h–j. The d-band centre and Bader charge of 3d metal-anchored C₂N have a strong linear relationship with the corresponding adsorption energy (R² is 0.88 and 0.96, respectively). However, the correlation of the d-band centre and Bader charge with adsorption energy gradually decreases with increasing metal period. The correlation coefficient (R²) values of the d-band centre and adsorption energy for 3d, 4d, and 5d transition metal anchored C₂N structures are 0.88, 0.75, and 0.72, respectively. The R² values of the Bader charge and adsorption energy for 3d, 4d, and 5d transition metal anchored C₂N structures are 0.96, 0.73, and 0.68, respectively. Thus, the d-band centre and Bader charge are potential descriptors of HCHO adsorption for 3d metals, but are not as satisfactory for 4d and 5d metals.

The electron localization function (ELF) was employed to further reveal the interactions between Hf-C₂N, Hf₂-C₂N and HCHO. (Fig. 3a and b) ELF values close to 0 indicate ionic bonds characterised by completely delocalised electrons, 0.5 indicates metallic bonds with electron gas-like behaviour and 1.0 indicates covalent bonds characterised by complete electron localisation. The ELF of Hf-C₂N and Hf₂-C₂N revealed that Hf has a metal bond with C with electron gas-like behaviour. This further illustrates Hf as an adsorption active site for Hf-C₂N and Hf₂-C₂N. Notably, the distance between the C and O atoms of HCHO was longer on Hf₂-C₂N than on Hf-C₂N (1.455 Å vs. 1.428 Å), suggesting that HCHO adsorption on Hf₂-C₂N was more activated and favoured subsequent degradation reactions.

Fig. 3 (a and b) The electron localization function of Hf-C₂N@HCHO and Hf₂-C₂N@HCHO. (c) The charge density difference of Hf-C₂N, Hf₂-C₂N, Hf-C₂N@HCHO and Hf₂-C₂N@HCHO, respectively. The isosurface value is 0.003 e Å⁻³.

To further explore the charge density transfer, we calculated the Bader charge and CDD (Fig. 3c) for Hf-C₂N and Hf₂-C₂N, with cyan representing charge dissipation and yellow for charge aggregation. In both Hf-C₂N and Hf₂-C₂N, significant charge transfer occurs from the introduced Hf atom to the C₂N substrate. The Hf atom of Hf-C₂N exhibits electron transfer and interaction mainly with the three N atoms on the surrounding C₂N pore. Each Hf atom of Hf₂-C₂N exhibts electron transfer with the two N atoms closer to the C₂N pore. A higher amount of electron transfer occurred in Hf₂-C₂N (2.663 |e⁻|) compared with Hf-C₂N (1.744 |e⁻|), indicating the stronger binding ability of Hf₂ and C₂N. When Hf-C₂N and Hf₂-C₂N adsorbed HCHO, electrons were mainly transferred from the Hf active site to HCHO, with a greater amount of electron transfer by Hf₂-C₂N (1.013 |e⁻|) relative to Hf-C₂N (0.805 |e⁻|) showing a stronger adsorption capacity for HCHO.

Projected density of states (PDOS) analysis was applied to gain deeper insights into the enhanced adsorption mechanism of HCHO by Hf-C₂N and Hf₂-C₂N, with a focus on orbital interactions. As shown in Fig. 4a–c, for Hf-C₂N and Hf₂-C₂N, Hf d is clearly hybridized with N p for the substrate C₂N above the Fermi level, suggesting that Hf has robust orbital coupling with C₂N, which favours the stability of Hf-C₂N and Hf₂-C₂N. The PDOS of HCHO adsorbed by Hf₂-C₂N is shown in Fig. 4c, Hf₂-C₂N@HCHO possesses abundant states around the Fermi level suggesting a superior adsorption activity of Hf₂-C₂N for HCHO.⁴⁴ The coupling of the Hf d to O p and C p orbitals of HCHO illustrated Hf as the active site for Hf₂-C₂N. The d-band centre developed by Nørskov was employed to unravel the mechanism behind the superior adsorption capacity of Hf₂-C₂N.⁴⁵ As a measure of the possible transition metal reactions, generally the increase of the d-band centre results in strong adsorption interaction.⁴⁶ As shown in Fig. 4b and c, for Hf₂-C₂N, after adsorption of HCHO, its d-band centre increased from 1.338 to 1.354 eV, indicating that Hf₂-C₂N favors a strong adsorption interaction. The cause of the raised Hf d of Hf₂-C₂N is presented in Fig. 4d; after adsorption of HCHO, the bandwidth of Hf d is shrunken, inducing an increasing level of the d-band centre, therefore resulting in a strong adsorption of HCHO by Hf₂-C₂N.

Fig. 4 (a and b) (a–c) The projected density of states of Hf-C₂N, Hf₂-C₂N and Hf₂-C₂N@HCHO, respectively. (d) Schematic diagram of the d-band centre bandwidth of Hf₂-C₂N before and after adsorption of HCHO. (e and f) The crystal orbital Hamiltonian population of C–O of HCHO for Hf-C₂N@HCHO and Hf₂-C₂N@HCHO, respectively.

The crystal orbital Hamiltonian population (COHP) was used to further expose the bonding and anti-bonding states between C and O atoms of HCHO. As shown in Fig. 4e and f, negative values of –pCOHP represent anti-bonding and positive values represent bonding. Near the Fermi energy level, the –pCOHP values of C and O atoms of HCHO adsorbed on Hf-C₂N are positive, indicating that HCHO is not sufficiently activated due to the filling of the bonding state near the Fermi energy level. However, the negative value of –pCOHP for HCHO adsorption on Hf₂-C₂N indicates that the antibonding state is populated around the Fermi energy level and that HCHO is sufficiently activated to favour the ensuing degradation. By integrating the energy up to the Fermi level, we obtain the ICOHP, which serves as a quantitative indicator. Generally, the more negative the ICOHP, the greater the combining strength.⁴⁷ The ICOHP of the C and O atoms of HCHO adsorbed to Hf₂-C₂N (−4.25) is more positive than that of Hf-C₂N (−4.83), indicating that the C and O atoms of HCHO interact more weakly, which facilitates the subsequent HCHO degradation reaction. The results are consistent with the previous results for PDOS, d-band centre and ELF.

In summary, following the screening of twenty-seven transition metals for stability and adsorption capacity in SA-C₂N and DA-C₂N, we have identified Hf₂-C₂N as a highly promising candidate with significant potential for HCHO adsorption. The electronic structure and orbital analysis revealed that the superior adsorption of HCHO on Hf₂-C₂N can be attributed to the additional introduction of Hf atoms, which enhances their activity. Additionally, we demonstrated that Hf₂-C₂N is more favorable for the subsequent degradation reaction of HCHO.

3.3 Possible catalytic degradation of HCHO

As described above, we carried out a systematic screening of SA-C₂N and DA-C₂N for the adsorption of HCHO and revealed that Hf₂-C₂N showed strong adsorption of HCHO and could be potentially used as a HCHO sensing material. Hydroxyl (˙OH) and superoxide (O₂˙⁻) radicals are essential in the degradation of HCHO.⁴⁸ As sources of ˙OH and O₂˙⁻, H₂O and O₂, it is worthwhile to study their adsorption on Hf₂-C₂N. Thus, we simulated the adsorption of H₂O and O₂ on Hf₂-C₂N in different configurations, respectively. As shown in Fig. 5a, the original configurations are presented on the left side of the figures, while the optimized configurations are displayed on the right side. It was noticeable that the two adsorption configurations of H₂O were induced by the strong adsorption of Hf₂-C₂N to form an –OH group and H atom, and the –OH group was bonded to two Hf atoms. Similarly, the strong adsorption of Hf₂-C₂N caused the O=O bond of both adsorbed configurations of O₂ to break, forming two O atoms chemically bonded to Hf atoms, and one of the O atoms is in between the two Hf atoms. Thus, the strong adsorption capacity of Hf₂-C₂N facilitates the adsorption of H₂O and the generation of hydroxyl radicals. On the other hand, Hf₂-C₂N also shows a high potential to activate O₂ molecules into superoxide radicals, thus facilitating the ensuing degradation reactions.

Fig. 5 (a) Before and after optimisation of configurations of H₂O and O₂ adsorption on Hf₂-C₂N and (b) Fukui index for intermediates of L–H and E–R mechanisms. (c) Potential energy profile of HCHO degradation in L–H and E–R mechanisms. (d) Potential energy profile of H₂O and CO₂ desorption.

The primary pathways for the catalytic oxidation of HCHO involve the Langmuir–Hinshelwood (L–H) and Eley–Rideal (E–R) mechanisms. The L–H mechanism involves HCHO adsorbing on the catalyst surface and reacting with a nearby reactive oxygen molecule to decompose into H₂O and CO₂, while the E–R mechanism involves HCHO reacting with O₂ on the catalyst surface to decompose into H₂O and CO₂.⁴⁹ To further investigate the reaction mechanism, we calculated the Fukui function f₀ of HCHO and O₂ for intermediates (IMs) in the process of the initial state (IS) and final state (FS) in both L–H and E–R mechanisms. In general, the higher the value of f₀, the more inclined to the radical reaction.^50–52 As shown in Fig. 5b, IM2, IM3, IM4, and IM5 were the share intermediates for the L–H and E–R mechanisms. For IM1 to IM4 intermediates, the O atom from O₂ had higher f₀ values, indicating that O atoms were more inclined to radical reactions. For IM5, the f₀ value of the H atom was higher, indicating that H₂O was more favourable to the formation of –OH and a H atom. Furthermore, we investigated the transition state (TS) of the intermediates for both L–H and E–R mechanisms; as shown in Fig. 5c, the maximum potential barrier for both the L–H and E–R mechanisms was the step IM4 → TS4 with a potential barrier of 2.1 eV, which was lower than that in previously reported theoretical studies for a potential Pt/TiO₂ catalyst (230.45 kJ mol⁻¹).⁵³ In the step IM5 → FS (Fig. 5d), CO₂ tended to desorb from the catalyst; however H₂O preferred to decompose into –OH and a H atom due to the lower potential barrier (1.56 eV vs. 7.82 eV). The production of –OH radicals was more favourable for the subsequent degradation of HCHO. In summary, Hf₂-C₂N is a potential catalyst for the degradation of HCHO due to its strong adsorption capacity which is more favourable to the formation of –OH and superoxide and the low reaction potential (2.1 eV) in the catalytic HCHO degradation steps.

The stability of the structure is an essential indicator of the synthesis and performance of the catalyst. To further investigate the potential ability of Hf₂-C₂N as the catalyst for HCHO detection adsorption and degradation, ab initio molecular dynamics (AIMD) simulations were subsequently performed for Hf₂-C₂N at 300 K, 500 K and 1000 K, respectively. As shown in Fig. 6, it was found that Hf₂-C₂N maintained its structure throughout and was not significantly damaged. The C₂N substrate structure of Hf₂-C₂N remained stable with some fluctuations during 6 ps. The detailed configuration structures are shown in Fig. S1.† Two Hf atoms were shifted in the pores of C₂N during 6 ps, and finally two Hf atoms protruded from the surface of C₂N, and the protruding Hf was favorable for the adsorption of HCHO.

Fig. 6 The potential energy of Hf₂-C₂N at 300 K, 500 K and 1000 K for 6000 fs.

4. Conclusion

In summary, we conducted a systematic density functional theory screening of twenty-seven transition metal single atom (SA) and double atom (DA) candidates for highly efficient formaldehyde adsorption. Among them, Hf₂-C₂N exhibited superior HCHO adsorption capacity with a more negative adsorption energy (−7.32 eV). Through various electronic structure and orbital analyses including partial density of states, d-band centre, CDD, Bader charge analysis, and ELF, we discovered that the introduction of an additional Hf atom into C₂N enhanced the activity of Hf atoms, resulting in strong HCHO adsorption activity. Furthermore, we investigated the degradation mechanism of HCHO on Hf₂-C₂N and found that it can potentially be degraded through strong adsorption interactions with water molecules and oxygen molecules to generate hydroxyl radicals (–OH) and superoxide radicals. The low reaction potential (2.1 eV) further supports the capability of Hf₂-C₂N for efficient degradation of HCHO. This study screens a promising diatomic material for highly efficient sensing and degradation while providing valuable insights for future experimental exploration.

Author contributions

Zhouhao Zhu: conceptualization, simulation, data analysis, writing – original draft. Hengcong Tao: supervision, funding acquisition. Renkun Zhang: supervision, writing – review & editing. Liyong Gan: supervision, writing – review & editing. Yingtang Zhou: conceptualization, supervision, writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the “Pioneer” and “Leading Goose” R&D Program of Zhejiang (2023C01191) and the National Natural Science Foundation of China (no. 22005269 and 22262024).

References

1. J. van den Broek, D. Klein Cerrejon, S. E. Pratsinis and A. T. Güntner, J. Hazard. Mater., 2020, 399, 123052.
2. C. Lou, G. Lei, X. Liu, J. Xie, Z. Li, W. Zheng, N. Goel, M. Kumar and J. Zhang, Coord. Chem. Rev., 2022, 452, 214280.
3. J. Bellat, I. Bezverkhyy, G. Weber, S. Royer, R. Averlant, J.-M. Giraudon and J.-F. Lamonier, J. Hazard. Mater., 2015, 300, 711–717.
4. P.-R. Chung, C.-T. Tzeng, M.-T. Ke and C.-Y. Lee, Sensors, 2013, 13, 4468–4484.
5. Y. Xu, L. Zheng, C. Yang, W. Zheng, X. Liu and J. Zhang, ACS Appl. Mater. Interfaces, 2020, 12, 20704–20713.
6. Y. Xu, C. Lou, L. Zheng, W. Zheng, X. Liu, M. Kumar and J. Zhang, Sens. Actuators, B, 2020, 307, 127616.
7. D. Wang, Z. Li, J. Zhou, H. Fang, X. He, P. Jena, J.-B. Zeng and W.-N. Wang, Nano-Micro Lett., 2017, 10, 4.
8. X. Li, X. Li, J. Wang and S. Lin, Sens. Actuators, B, 2015, 219, 158–163.
9. D. Zhang, Y. Cao, Z. Yang and J. Wu, ACS Appl. Mater. Interfaces, 2020, 12, 11979–11989.
10. G. Zhang, S. Zhang, L. Yang, Z. Zou, D. Zeng and C. Xie, Sens. Actuators, B, 2013, 188, 137–146.
11. A. Dey, Mater. Sci. Eng. B, 2018, 229, 206–217.
12. Y. Ravi Kumar, K. Deshmukh, T. Kovářík and S. K. Khadheer Pasha, Coord. Chem. Rev., 2022, 461, 214502.
13. Y. Zhou, H. Yin and S. Ai, Coord. Chem. Rev., 2021, 447, 214156.
14. Y. Chen, S. Ji, C. Chen, Q. Peng, D. Wang and Y. Li, Joule, 2018, 2, 1242–1264.
15. R. Riaz, M. Ali, I. A. Sahito, A. A. Arbab, T. Maiyalagan, A. S. Anjum, M. J. Ko and S. H. Jeong, Appl. Surf. Sci., 2019, 480, 1035–1046.
16. K. Wang, L. Jiang, T. Xin, Y. Li, X. Wu and G. Zhang, Chem. Eng. J., 2022, 429, 132229.
17. H. Shin, W.-G. Jung, D.-H. Kim, J.-S. Jang, Y. H. Kim, W.-T. Koo, J. Bae, C. Park, S.-H. Cho, B. J. Kim and I.-D. Kim, ACS Nano, 2020, 14, 11394–11405.
18. M. Chen, H. Wang, J. Wang, M. Sun, Y. Hu, X. Zhao and Y. Zhou, Phys. Chem. Chem. Phys., 2023, 25, 25353–25360.
19. X. Zhao, J. Chen, Z. Bi, S. Chen, L. Feng, X. Zhou, H. Zhang, Y. Zhou, T. Wågberg and G. Hu, Adv. Sci., 2023, 10, 2205889.
20. J. Fu, J. Dong, R. Si, K. Sun, J. Zhang, M. Li, N. Yu, B. Zhang, M. G. Humphrey, Q. Fu and J. Huang, ACS Catal., 2021, 11, 1952–1961.
21. B. Kim, D. Kwon, J.-O. Baeg, M. Austeria P, G. H. Gu, J.-H. Lee, J. Jeong, W. Kim and W. Choi, Adv. Funct. Mater., 2023, 33, 2212453.
22. X. Guo, J. Gu, S. Lin, S. Zhang, Z. Chen and S. Huang, J. Am. Chem. Soc., 2020, 142, 5709–5721.
23. H. Zhang, G. Guo, Z. Wang, Q. He, X. He and H. Ji, Appl. Catal., B, 2023, 333, 122774.
24. Z. Liu, J. Wei, G. Zhang, D. Zhang, J. Zhang, W. Yang, C. Wu and I. D. Gates, J. Mater. Chem. A, 2023, 11, 1978–1990.
25. H. Tang, J. Zhang, M. Huang, J. Zhang, Y. Zhou, G. Wang, R. Wang and J. Chen, J. Colloid Interface Sci., 2022, 624, 527–536.
26. J. Mahmood, E. K. Lee, M. Jung, D. Shin, I.-Y. Jeon, S.-M. Jung, H.-J. Choi, J.-M. Seo, S.-Y. Bae, S.-D. Sohn, N. Park, J. H. Oh, H.-J. Shin and J.-B. Baek, Nat. Commun., 2015, 6, 6486.
27. L. Zhu, Q. Xue, X. Li, T. Wu, Y. Jin and W. Xing, J. Mater. Chem. A, 2015, 3, 21351–21356.
28. W. Zhong, Y. Qiu, H. Shen, X. Wang, J. Yuan, C. Jia, S. Bi and J. Jiang, J. Am. Chem. Soc., 2021, 143, 4405–4413.
29. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
30. X. Zhao, M. Chen, Z. Bi, H. Zhang, G. Hu and Y. Zhou, Small, 2023, 8, 2304854.
31. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 154104.
32. P. E. Blöchl, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 17953–17979.
33. V. Wang, N. Xu, J.-C. Liu, G. Tang and W.-T. Geng, Comput. Phys. Commun., 2021, 267, 108033.
34. G. Henkelman and H. Jónsson, J. Chem. Phys., 2000, 113, 9978–9985.
35. Y. Yang, M. Guo, G. Zhang and W. Li, Carbon, 2017, 117, 120–125.
36. J. Mahmood, E. K. Lee, M. Jung, D. Shin, I.-Y. Jeon, S.-M. Jung, H.-J. Choi, J.-M. Seo, S.-Y. Bae, S.-D. Sohn, N. Park, J. H. Oh, H.-J. Shin and J.-B. Baek, Nat. Commun., 2015, 6, 6486.
37. J. Du, C. Xia, W. Xiong, X. Zhao, T. Wang and Y. Jia, Phys. Chem. Chem. Phys., 2016, 18, 22678–22686.
38. P. Janthon, S. M. Kozlov, F. Viñes, J. Limtrakul and F. Illas, J. Chem. Theory Comput., 2013, 9, 1631–1640.
39. H. Liu, X. Wang, C. Pan and K. M. Liew, J. Phys. Chem. C, 2012, 116, 8044–8053.
40. Z. Zhu, H. Tao, J. Fu, Y. Zhou, J. Guo and C. Zhai, Chin. Chem. Lett., 2023, 34, 107476.
41. T. Jia, L. Wang, Z. Zhu, B. Zhu, Y. Zhou, G. Zhu, M. Zhu and H. Tao, Chin. Chem. Lett., 2023, 108692,  DOI:10.1016/j.cclet.2023.108692.
42. D. Mei, N. Aaron Deskins and M. Dupuis, Surf. Sci., 2007, 601, 4993–5001.
43. Z. Zhu, M. Chen, M. Sun, J. Wang, Y. Zhou, X. Li and H. Tao, Phys. Chem. Chem. Phys., 2022, 24, 26776–26784.
44. J. Wang, Y.-P. Liu, H. Zhang, D.-J. Huang and K. Chu, Catal. Sci. Technol., 2019, 9, 4248–4254.
45. M. Mavrikakis, B. Hammer and J. K. Nørskov, Phys. Rev. Lett., 1998, 81, 2819–2822.
46. X. Liao, R. Lu, L. Xia, Q. Liu, H. Wang, K. Zhao, Z. Wang and Y. Zhao, Energy Environ. Mater., 2022, 5, 157–185.
47. X. Guo, J. Gu, X. Hu, S. Zhang, Z. Chen and S. Huang, Catal. Today, 2020, 350, 91–99.
48. Y. Su, W. Li, G. Li, Z. Ao and T. An, Chin. J. Catal., 2019, 40, 664–672.
49. C. Mang, G. Li, M. Rao, X. Zhang, J. Luo and T. Jiang, Appl. Surf. Sci., 2023, 608, 154964.
50. Y. Ma, J. Liang, D. Zhao, Y.-L. Chen, J. Shen and B. Xiong, RSC Adv., 2014, 4, 17262–17264.
51. Z. Nie, M. Chen, L. Zhang, Q. Feng, J. Hu, X. Huang, C. Zhou, Y. Zhou, T. Wågberg and G. Hu, Chem. Eng. J., 2023, 463, 142411.
52. J. Wang, M. Sun, C. Liu, Y. Ye, M. Chen, Z. Zhao, Y. Zhang, X. Wu, K. Wang and Y. Zhou, Adv. Mater., 2023, 35, 2306103.
53. J. Ding, Y. Yang, J. Liu and Z. Wang, Chemosphere, 2020, 248, 125980.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3ta06132a

---