Infrared Photodissociation Spectroscopy of Be(N₂) _n ⁺ and Be(CO) _n ⁺ (n = 3–4): Charge Effects on Bonding and Coordination

Abstract

Beryllium dinitrogen and carbonyl cation complexes, Be(N₂) _n ⁺ and Be(CO) _n ⁺ (n = 3–4), were produced in the gas phase by a laser vaporization‐supersonic expansion ion source and investigated by infrared photodissociation spectroscopy in conjunction with quantum chemical calculations. Be(N₂)₃ ⁺ is assigned to a planar D _3h‐symmetric structure with three equivalent end‐on coordinated molecular N₂ ligands, whereas Be(N₂)₄ ⁺ is characterized as a weakly bound N₂ adduct of Be(N₂)₃ ⁺. The corresponding carbonyl complexes exhibit analogous structural motifs. Bonding analyses reveal that the L → Be (L = NN/CO) interactions in these end‐on complexes are governed by L → Be σ donation, accompanied by weaker L ← Be π back‐donation. In contrast to our previous study on Be(N₂)₃ (Angew. Chem. Int. Ed. 2020, 59, 10603), where both end‐on and side‐on coordination motifs coexist, only end‐on structures are observed for the cationic species. The absence of side‐on coordination is attributed to the positive charge, which significantly weakens π back‐donation while only modestly enhancing σ donation, thereby destabilizing side‐on binding motifs.

Gas‐phase Be(N₂)₃ ⁺ is characterized via infrared photodissociation spectroscopy. Positive charge suppresses Be⁺ → N₂ π back donation, destabilizing side‐on coordination, thereby leading to a preference for end‐on coordination.

1. Introduction

Activation and functionalization of dinitrogen (N₂) remain challenging in heterogeneous catalysis and coordination chemistry [1, 4]. A fundamental question is how active sites interact with N₂ and weaken the exceptionally strong N≡N triple bond. Such metal–ligand interactions are commonly described by the Dewar–Chatt–Duncanson (DCD) bonding model [5], which involves ligand‐to‐metal σ donation accompanied by metal‐to‐ligand π back‐donation [6, 8]. In conventional transition metal complexes, low‐lying valence d orbitals enable efficient π back‐donation into ligand antibonding orbitals, thereby weakening the N≡N bond and promoting activation. These interactions also give rise to diverse coordination modes [2, 9, 10]. End‐on coordination is generally dominant, whereas side‐on coordination, although often associated with stronger bond activation, remains comparatively rare.

For main group elements, side‐on coordination of N₂ is even less common because of the absence of suitably low‐lying d orbitals required for effective π back‐donation. Consequently, only a limited number of side‐on dinitrogen complexes of main group elements have been identified, typically under low‐temperature matrix‐isolation or gas‐phase conditions [11, 17]. Among main‐group elements, beryllium is particularly intriguing because its 2p orbitals can participate in π‐type bonding interactions, enabling unusual chemical bonds, and in some systems are involved in tunneling‐mediated reaction dynamics [18, 22]. Previous studies have demonstrated that neutral beryllium species can stabilize side‐on coordinated N₂ ligands with substantial bond activation through an excited 1s²2s⁰2p² electronic configuration [16, 17, 23], giving rise to a p‐orbital analog of DCD‐type synergistic bonding [24]. In this bonding scheme, vacant Be 2s, 2p _x , and 2p _y orbitals accept ligand σ donation, while an occupied 2p _z orbital enables π back‐donation into ligand antibonding orbitals. However, neutral beryllium complexes often exhibit closely lying electronic states and delicate energetic competition between end‐on and side‐on coordination geometries [17], posing significant challenges for reliable quantum chemical calculations.

In contrast, cationic beryllium complexes are expected to exhibit reduced electronic near‐degeneracy and a simplified electronic structure, while also potentially altering coordination references [19, 25, 26]. Infrared photodissociation (IRPD) spectroscopy has revealed an unexpected side‐on bridging CO ligand in the homoleptic dinuclear carbonyl cation Be₂(CO)₅ ⁺, which is otherwise typically assumed to contain only terminal CO ligands [26]. However, it remains unclear whether side‐on coordinated N₂ or CO ligands can also be stabilized in mononuclear beryllium cation complexes.

Herein, we combine IRPD spectroscopy with quantum chemical calculations to investigate mononuclear beryllium cations Be(NN) _n ⁺ (n = 3–4), using the corresponding carbonyl complexes Be(CO) _n ⁺ (n = 3–4) as reference systems. By comparing Be⁺─NN and Be⁺─CO bonding, we aim to elucidate the nature of their interactions and to clarify how positive charge influences Be–ligand bonding and coordination behavior, with particular emphasis on the competition between end‐on and side‐on coordination motifs.

2. Results and Discussion

Typical mass spectra of cationic beryllium complexes generated by laser vaporization of a beryllium target in N₂‐ and CO‐seeded supersonic expansions are shown in Figures S1, S2, respectively. In both systems, the mononuclear cationic complexes Be(NN) _n ⁺ and Be(CO) _n ⁺ dominate the mass spectra, with well‐resolved peak series corresponding to n = 3–10. Cation complexes with n = 3–4, which are the focus of the present work, were selected for IRPD spectroscopy. Upon resonant excitation of an IR‐active vibrational mode, the cations absorb photons and undergo fragmentation. The vibrational spectra are obtained by monitoring the fragment yield as a function of the IR laser wavenumber.

2.1. Be(NN) _n ⁺

The IRPD spectra of the Be(NN) _n ⁺ (n = 3–4) are shown in Figure 1. Both complexes exhibit strong absorption bands in the 2060–2140 cm⁻¹ region, corresponding to N≡N stretching vibrations of coordinated dinitrogen ligands. No detectable IRPD signals were observed in the 1700–1950 cm⁻¹ region; therefore, the spectral range is not shown. For Be(NN)₃ ⁺, a broad band centered at 2109 cm⁻¹ is observed (Figure 1a) and is tentatively assigned to the doubly degenerate N≡N stretching mode of a D _3h‐symmetric structure with three equivalent end‐on coordinated N₂, consistent with similar spectral features reported for neutral Be(NN)₃ in solid neon [17]. The photodissociation yield is low (∼6%), indicating a strong Be⁺–NN interaction.

FIGURE 1.

IRPD spectra of the Be(NN) _n ⁺ (n = 3–4) complexes and simulated harmonic spectra of low‐energy isomers calculated using B3LYP‐D3(BJ)/aug‐cc‐pVTZ. A scaling factor of 0.952 was applied to the calculated frequencies. Relative energies (kcal mol⁻¹) were calculated using CCSD(T)/aug‐cc‐pVTZ including zero‐point vibrational energy (ZPE) corrections.

To support the assignment, quantum chemical calculations were carried out to optimize structures and to calculate the infrared spectra of Be(NN)₃ ⁺ at the B3LYP‐D3(BJ)/aug‐cc‐pVTZ level. The optimized geometries are given in Figure 2. The global minimum, Be(η ¹‐N₂)₃ ⁺ (denoted as (NN)₃‐A), adopts a planar D _3h‐symmetric doublet structure, in which three N₂ ligands are coordinated end‐on to the central Be atom with uniform Be─N bond lengths of 1.622 Å. A higher energy isomer, (NN)₂Be(η ²‐N₂)⁺ (denoted as (NN)₃‐B), is calculated to lie 8.8 kcal mol⁻¹ above (NN)₃‐A at the CCSD(T)/aug‐cc‐pVTZ level. This isomer has a C _2v‐symmetric structure and contains one side‐on bound N₂ ligand together with two equivalent end‐on η ¹‐N₂ ligands (Figure 2). The side‐on N₂ is characterized by two equivalent Be─N bonds of 1.740 Å. The calculated harmonic spectra of these two isomers are compared with the experimental IRPD spectrum in Figure 1. (NN)₃‐A reproduces the observed feature at ∼2100 cm⁻¹, whereas (NN)₃‐B exhibits a strong band below 2000 cm⁻¹ and weak bands near 2250 cm⁻¹ that are absent in the experiment. These results unambiguously identify (NN)₃‐A as the observed structure.

FIGURE 2.

Low energy structures of Be(NN) _n ⁺ and Be(CO) _n ⁺ (n = 3–4) calculated using B3LYP‐D3(BJ)/aug‐cc‐pVTZ method. Selected bond distances (Å) and bond angles (°) are given in Figure. D _e denotes the bond dissociation energy (kcal mol⁻¹) for the process Be(L) _n ⁺ → Be⁺ + n L (L = NN, CO).

As shown in Figure 1, the IRPD spectrum of Be(NN)₄ ⁺ exhibits two resolved absorption features in the N≡N stretching region, with a dominant band centered at 2115 cm⁻¹ and a shoulder at 2099 cm⁻¹. Quantum chemical calculations indicate that the lowest energy structure consists of a Be(η ¹‐N₂)₃ ⁺ core weakly tagged by an additional N₂ ligand (Figure 2). The simulated IR spectrum of Be(NN)₄ ⁺ reproduces well the experimentally observed band splitting. The experimental band positions and calculated vibrational frequencies are listed in Table 1.

TABLE 1.

Observed band positions (cm⁻¹), calculated harmonic vibrational frequencies (scaled by factors of 0.952 and 0.971 for Be(NN) _n ⁺ and Be(CO) _n ⁺, respectively), and relative IR intensities for Be(NN) _n ⁺ and Be(CO) _n ⁺ (n = 3–4).

	Be(NN) n +			Be(CO) n +
	Exptl.	Calcd.		Exptl.	Calcd.
		(NN) n ‐A	(NN) n ‐B		(CO) n ‐A
n = 3	2109	2100 (1015)	1895 (641)	2112	2087 (1153)
			2231 (179)
			2277 (86)
n = 4	2099	2098 (1105)		2128	2082 (1173)
	2115	2109 (974)		2176	2145 (50)

Compared with previous reports on neutral Be(N₂)₃ [17], a pronounced change in coordination preference is observed upon the cationic formation in Be(N₂)₃ ⁺. Single‐point energy calculations at the CCSD(T)/aug‐cc‐pVTZ level show that, for Be(N₂)₃, the end‐on and side‐on isomers differ by only 1.3 kcal mol⁻¹ (Table 2), with the side‐on structure being slightly more stable. This small energy gap allows both coordination modes to coexist under low‐temperature conditions. In contrast, for the cationic species Be(N₂)₃ ⁺, the energy difference between the two isomers increases to 8.8 kcal mol⁻¹ (Table 2), with the end‐on structure becoming clearly favored. This substantial stabilization of the end‐on isomer renders the side‐on coordination inaccessible under the present condition.

TABLE 2.

EDA‐NOCV results of Be(NN)₃, Be(NN)₃ ⁺, and Be(CO)₃ ⁺, obtained using the interaction fragments Be(NN)₂ (singlet, S) and N₂ (S) for the neutral system, and Be(NN/CO)₂ ⁺ (doublet, D) and N₂/CO (S) for the cationic system.

Energy terms	Assignment	Be(NN) 3		Be(NN) 3 +		Be(CO) 3 +
		Be(NN)2 (S) + (η 1‐N2) (S)	Be(NN)2 (S) + (η 2‐ N2) (S)	Be(NN)2 + (D) + (η 1‐N2) (S)	Be(NN)2 + (D) + (η 2‐N2) (S)	Be(CO)2 + (D) + CO (S)
Relative energies		1.3	0.0	0.0	8.8
ΔE int		−48.1	−56.1	−44.3	−38.1	−50.6
ΔE Pauli		42.6	71.0	34.3	47.3	37.9
ΔE disp		−2.3	−2.7	−2.2	−2.6	−2.1
ΔE elstat a		−24.1 (27.3%)	−14.2 (11.4%)	−21.6 (28.3%)	−0.1 (0.1%)	−32.3 (37.4%)
ΔE orb a		−64.3 (72.7%)	−110.2 (88.6%)	−54.8 (71.7%)	−82.7 (99.9%)	−54.1 (62.6%)
ΔE orb(1) b	Be(NN/CO)2 0/+	−31.6 (49.1%)	−77.5 (70.3%)	−10.1 (18.4%)	−35.5 (42.9%)	−10.6 (19.6%)
	→ N2/CO
	π back‐donation
ΔE orb(2) b	Be(NN/CO)2 0/+	−24.3 (37.8%)	−23.1 (21.0%)	−28.8 (52.6%)	−29.3 (35.4%)	−35.1 (64.9%)
	← N2/CO
	σ donation
ΔE orb(3) b	polarization	−5.3 (8.2%)	−8.6 (7.8%)	−8.8 (16.1%)	−12.1 (14.6%)	−6.5 (12.0%)
ΔE orb(rest) b		−3.1 (4.9%)	−1.0 (0.9%)	−7.1 (12.9%)	−5.8 (7.1%)	−1.9 (3.5%)

Note: The calculations were performed at the B3LYP‐D3(BJ)/TZ2P//B3LYP‐D3(BJ)/aug‐cc‐pVTZ level. Relative energies were calculated at the CCSD(T)/aug‐cc‐pVTZ//B3LYP‐D3(BJ)/aug‐cc‐pVTZ level. Energy values are in kcal mol⁻¹.

^a The values in the parentheses show the contribution to the total attractive interaction ΔE _elstat plus ΔE _orb.

^b The values in the parentheses show the contribution to the total orbital interaction ΔE _orb.

2.2. Be(CO) _n ⁺

The IRPD spectra of Be(CO) _n ⁺ (n = 3–4) are shown in Figure 3. For both complexes, the dominant absorption appears near 2120 cm⁻¹ and is assigned to the C≡O stretching vibration. The IRPD spectrum of Be(CO)₃ ⁺ exhibits a broad band centered at 2112 cm⁻¹ with a low fragmentation yield (∼1%), indicating a strong Be⁺–CO interaction. Theoretical calculations using B3LYP‐D3(BJ)/aug‐cc‐pVTZ predict a planar D _3h‐symmetric doublet structure (Figure 2), consisting of three end‐on coordinated CO ligands with an equilibrium Be─CO bond length of 1.711 Å. The IRPD spectrum of Be(CO)₄ ⁺ displays an additional weak band at 2176 cm⁻¹, indicating the presence of a weakly bound CO ligand [8, 27, 29]. Accordingly, theoretical calculations suggest that the lowest‐energy structure of Be(CO)₄ ⁺ consists of a D _3h‐symmetric Be(CO)₃ ⁺ core tagged by an additional CO molecule (Figure 2). Such structural motifs of Be(CO) _n ⁺ (n = 3, 4) are highly similar to those found for the corresponding Be(NN) _n ⁺. As shown in Figure 3, the simulated spectra are in excellent agreement with the experimental observations for Be(CO) _n ⁺ (n = 3–4).

FIGURE 3.

IRPD spectra of the Be(CO) _n ⁺ (n = 3–4) compared with simulated harmonic spectra calculated using B3LYP‐D3(BJ)/aug‐cc‐pVTZ. A scaling factor of 0.971 was applied to calculated frequencies.

2.3. Bonding Analysis

Atoms‐in‐molecules (AIM) charge analysis within quantum topological framework [30, 31] was performed to examine whether the positive charge is localized on the beryllium atom or is delocalized over the ligands. Such partitions of the electron density into atomic basis provides a chemically meaningful insights in charge distribution, which has been widely used to characterize charge localization and charge–transfer effects [19, 32, 33]. The calculated AIM charge distributions of neutral Be(NN)₃ and cationic Be(NN)₃ ⁺ in both end‐on and side‐on coordination modes are summarized in Table S1. The results show that the positive charge on the central beryllium atom increases only slightly from Be(NN)₃ to Be(NN)₃ ⁺, indicating that the additional positive charge is delocalized over the Be(NN)₂ moiety rather than exclusively on the Be center. For both neutral and cationic complexes, the Be(NN)₂ moiety carries a higher positive charge in the side‐on geometries compared to the end‐on analogs. This trend suggests reduced electron donation from Be to the coordinated N₂ ligands in side‐on geometries relative to end‐on coordination.

To clarify the nature of this pronounced charge‐induced effect, energy decomposition analysis combined with natural orbitals for chemical valence (EDA‐NOCV) was employed to dissect the bonding interactions in these complexes. The calculations were performed at the B3LYP‐D3(BJ)/TZ2P level based on the B3LYP‐D3(BJ)/aug‐cc‐pVTZ optimized geometries. The results are summarized in Table 2. The bonding analyses were performed using Be(NN)₂/Be(NN)₂ ⁺ and N₂ fragments in their respective spin states to compare the end‐on (η ¹‐N₂) and side‐on (η ²‐N₂) coordination modes.

For the neutral complex Be(NN)₃, the intrinsic interaction energy (ΔE _int) of the side‐on coordinated isomer, (NN)₂Be(η ²‐N₂) (−56.1 kcal mol⁻¹), is larger than that of the end‐on isomer, (NN)₂Be(η ¹‐N₂) (−48.1 kcal mol⁻¹). This difference arises mainly from electrostatic (ΔE _elstat) and covalent (ΔE _orb) contributions, with orbital interactions dominating in both coordination modes. Decomposition of ΔE _orb into pairwise orbital interaction terms reveals that the dominant contribution, ΔE _orb(1), corresponds to Be(NN)₂ → N₂ π back‐donation, whereas the second major contribution, ΔE _orb(2), arises from N₂ → Be(NN)₂ σ donation. The third term, ΔE _orb(3), is comparatively small and associated with minor polarization interaction. The associated deformation densities (Δρ) are shown in Figure 4. Notably, the side‐on coordinated complex exhibits significantly enhanced π back‐donation, which accounts for its stronger overall interaction and greater covalent stabilization relative to the end‐on isomer.

FIGURE 4.

Shapes of the deformation densities, Δρ _{(1) − (2)}, which are associated with the two dominant pairwise orbital interaction terms ΔE _{orb(1) − (2)} in Be(η ¹‐N₂)₃, Be(η ¹‐N₂)₃ ⁺, (NN)₂Be(η ²‐N₂), (NN)₂Be(η ²‐N₂)⁺, and Be(CO)₃ ⁺. The isosurface values are 0.001 a.u. The charge flow of the deformation densities is red → blue. Energy values are in kcal mol⁻¹.

In contrast, a marked reversal of the bonding pattern is observed in the cationic complex Be(NN)₃ ⁺, as shown in Table 2. The ΔE _int is now larger in magnitude for the end‐on isomer Be(η ¹‐N₂)₃ ⁺ (−44.3 kcal mol⁻¹) than for the side‐on isomer (NN)₂Be(η ²‐N₂)⁺ (−38.1 kcal mol⁻¹). Significant differences are found in both electrostatic ΔE _elstat and orbital interaction ΔE _orb components. In particular, the side‐on η ²‐N₂ coordination shows a nearly zero electrostatic contribution, indicating a more pronounced covalent character of the Be⁺–(η ²–N₂) interactions; However, this loss of electrostatic stabilization ΔE _elstat weakens the overall interaction energy. The orbital interaction energy and the associated deformation densities (Figure 4) in Be(N₂)₃ ⁺ generally follow the trends observed for neutral Be(N₂)₃. However, the positive charge significantly suppresses Be(NN)₂ ⁺ → N₂ π back‐donation, while concomitantly enhancing N₂ → Be(NN)₂ ⁺ σ donation (Figure 4), particularly in the end‐on isomer. As a result, σ donation becomes the dominant orbital contribution in the end–on complex (ΔE _orb(2) = −28.8 kcal mol⁻¹), exceeding π back‐donation (ΔE _orb(1) = −10.1 kcal mol⁻¹). In the side‐on isomer, π back‐donation remains substantial (ΔE _orb(1) = −35.5 kcal mol⁻¹), but is only slightly larger than σ donation by ∼6.2 kcal mol⁻¹. The simultaneous weakening of electrostatic stabilization and π back‐donation interaction inevitably leads to a reduction in the intrinsic interaction energy of (NN)₂Be⁺–(η ²‐N₂). This electronic reorganization rationalizes the enhanced stability of the end‐on Be(NN)₃ ⁺ structure and the experimental observation.

Because CO is isoelectronic with N₂, further bonding analyses were conducted for the analogous Be(CO)₃ ⁺ system to clarify the nature of Be–ligand bonding. The dissociation energy (D _e) for the process Be(L) _n ⁺ → Be⁺ + n L (L = NN, CO) shows a stronger bonding in CO complexes than in N₂ complexes (Figure 2). The EDA‐NOCV results are summarized in Table 2, using Be(CO)₂ ⁺ and CO in their respective spin states as interacting fragments. The energy decomposition terms for Be(CO)₃ ⁺ are generally comparable to those of Be(NN)₃ ⁺, but Be(CO)₃ ⁺ exhibits a stronger overall interaction energy ΔE _int, mainly due to a larger electrostatic contribution ΔE _elstat for CO. The ΔE _elstat term increases from  −21.6 kcal mol⁻¹ in Be(NN)₃ ⁺ to  −32.3 kcal mol⁻¹. Although the orbital interaction energies are of similar magnitude in both complexes, σ donation (OC → Be(CO)₂ ⁺,  −35.1 kcal mol⁻¹) is stronger than that for η ¹‐N₂ (−28.8 kcal mol⁻¹). However, CO and N₂ exhibit comparable metal‐ligand π back‐donation contributions, with values of −10.1 kcal mol⁻¹ for N₂, and  −10.6 kcal mol⁻¹ for CO. This trend is consistent with previous studies on boron cluster complexes [12, 34] but is less commonly observed in conventional transition metal–ligand bonding. Overall, these results show that larger dissociation energies of the CO complexes arise primarily from enhanced σ donation and electrostatic stabilization, whereas the π–acceptor interactions of CO and N₂ with Be⁺ remain remarkably similar.

3. Conclusion

In summary, IRPD spectroscopy combined with theoretical calculations has been used to explore the structures and bonding motifs of Be(NN) _n ⁺ and Be(CO) _n ⁺ (n = 3–4) in the gas phase. The results demonstrate that end‐on coordination is favored in both systems. Be(NN)₃ ⁺ is identified as a planar D _3h‐symmetric structure with three equivalent end‐on bound N₂ ligands, whereas Be(NN)₄ ⁺ consists of a stable Be(NN)₃ ⁺ core weakly tagged by an additional N₂ ligand. Be(CO) _n ⁺ (n = 3–4) exhibit analogous structural motifs to the corresponding Be(NN) _n ⁺ systems. Bonding analyses reveal that the interaction between Be(NN)₂ ⁺ and η ¹‐N₂ in Be(NN)₃ ⁺ is governed by significant N₂ → Be(NN)₂ ⁺ σ donation accompanied by Be(NN)₂ ⁺ → N₂ π back‐donation. A similar donor–acceptor bonding pattern is found in Be(CO)₃ ⁺. Compared with the neutral system, the positive charge significantly reduces electrostatic interaction and suppresses Be⁺ → N₂ π back‐donation, which is crucial for stabilizing side‐on coordination. This electronic effect ultimately leads to the exclusive formation of end‐on bound structures in cationic beryllium complexes.

4. Experimental and Theoretical Methods

The cations were generated in the gas phase using a laser vaporization supersonic expansion ion source and characterized by IRPD spectroscopy with a collinear tandem time‐of‐flight mass spectrometer. The experimental setup has been described previously in detail [35]. A beryllium metal target was ablated using the fundamental output of a Nd:YAG laser (Continuum Minilite II) with a typical pulse energy of 10–20 mJ/pulse. The resulting plasma was entrained in a pulsed supersonic expansion of helium seeded with 5%–10% N₂ or CO using a pulsed valve (General Valve Series 9) operated at a backing pressure of 0.6–1.2 MPa. After supersonic expansion, the ion beams were collimated by a skimmer and subsequently extracted into a collinear tandem time‐of‐flight mass spectrometer. The ions of interest were mass‐selected and decelerated before interaction with a tunable infrared laser beam in the photodissociation region. The dissociation IR laser was generated from an optical parametric oscillator/amplifier system (LaserVision) pumped by a Nd:YAG laser (Continuum Surelite EX). The infrared pulse energy was typically in the range of 0.2–1.0 mJ/pulse over the spectral range investigated. After infrared irradiation, the fragment ions together with the undissociated parent ions were reaccelerated and detected using the second‐stage time‐of‐flight mass spectrometer. IRPD spectra were obtained by monitoring the wavelength‐dependent fragment ion yield. Spectra were recorded with a step size of 2 cm⁻¹ and averaged over approximately 350 shots at each wavenumber.

Quantum chemical calculations were performed for Be(NN) _n ⁺ and Be(CO) _n ⁺ (n = 3–4) using Gaussian 09 [36]. All structures were optimized using B3LYP‐D3(BJ) [37, 38] and aug‐cc‐pVTZ basis set [39, 40]. Single point energies were refined using CCSD(T) [41]/aug‐cc‐pVTZ. Calculated harmonic frequencies were scaled using two independent factors: 0.952 for Be(NN) _n ⁺ and 0.971 for Be(CO) _n ⁺. These scaling factors were derived from the experimental‐to‐calculated stretching frequencies ratio of N₂ (2330/2447 cm⁻¹) and CO (2143/2208 cm⁻¹), respectively. Simulated infrared spectra were generated using Lorentzian line‐shape broadening with a full width at half‐maximum of 8 cm⁻¹. AIM charge analyses were carried out using the Multiwfn package [42]. EDA‐NOCV [43, 44] was performed using the ADF 2026.103 program package to elucidate the nature of bonding interactions.

Funding

This study was supported by Natural Science Foundation of Guangxi Province (2026GXNSFBA00640313), Guangxi University (ZX01080030425011), National Natural Science Foundation of China (21873020), and Shanghai Pujiang Program (25PJA006).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Material

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.