Observation of double-ring tubular B₂₀(CO)_n⁺ (n = 1–8): emergence of 2D-to-3D transition in boron carbonyl complexes

Hong Niu, Qiang Chen*, Rui-Nan Yuan, Qin-Wei Zhang and Si-Dian Li*
Institute of Molecular Science, Shanxi University, Taiyuan 030006, P. R. China. E-mail: chenqiang@sxu.edu.cn; lisidian@sxu.edu.cn

First published on 25th July 2025

---

---

Introduction

Carbon monoxide (CO) is one of the most important ligands in chemistry due to its unique capacity to form σ-donation and π-back-donation coordination interactions in metal carbonyl complexes.^1–3 In typical transition-metal (TM) carbonyl complexes, the :CO ligand donates its σ-HOMO lone-pair on carbon to the partially occupied (n − 1)d orbitals of the TM center while concurrently accepting back donation from the (n − 1)d orbitals of the TM center in its two degenerate π*-LUMO antibonding orbitals. σ + π dual coordination interactions dominate the electronic structures and reactivities of TM carbonyl complexes. Main group metal carbonyl complexes have also been discovered in the past decade, including the cubic alkaline-earth metal carbonyl complexes M(CO)₈ (M = Ca, Sr, and Ba) and Be(CO)_n (n = 1–4), in which the alkaline-earth metals serve as “honorable transition metals” exhibiting transition-metal-like behaviors.^4,5

As a prototypical electron-deficient element in the periodic table, boron (He[2s²2p¹]) can also form various types of carbonyl complexes. The closed-shell carbonyl borane H₃BCO and its derivatives are experimentally known as stable boron carbonyl complexes.⁶ In recent years, various small boron carbonyl clusters, including linear BCO,⁷ OCBBCO,⁸ and BBCO,⁹ V-shaped B(CO)₂,¹⁰ rhombic B₄(CO)₂,¹¹ and planar (2D) B(CO)₃⁺, B(CO)₄⁺, B₂(CO)₄⁺, and B₃(CO)_n⁺ (n = 3–6), have been observed in experiments using infrared photodissociation (IRPD) spectroscopy.^12–15 Neutral rhombic B₄(CO)₃ was also observed in the gas phase.¹⁶ In these boron carbonyl complexes, CO ligands serve as donors to coordinate B_n centres as acceptors via effective σ-donations. Based on joint gas-phase mass spectroscopy, collision-induced dissociation (CID), and first-principles theory investigations, our group observed in 2024 the first quasi-planar boron carbonyl aromatics (BCAs) B₁₃(CO)_n⁺ (n = 1–7), analogous to benzene C₆H₆ under ambient conditions.¹⁷ CID experiments confirmed that CO ligands are molecularly coordinated to the aromatic B₁₃⁺ core in these BCAs without being activated or disassociated. Similarly, in 2025, quasi-planar B₁₁(CO)_n⁺ (n = 1–6) and B₁₅(CO)_n⁺ (n = 1–5) with π and σ conflicting aromaticity were observed in the gas phase, presenting the largest boron carbonyl complexes observed to date.¹⁸ Neutral double-ring tubular B₂₀(CO)_n (n = 1–8) complexes based on the framework of aromatic DR tubular B₂₀ have also been predicted in theory recently.¹⁹ However, to the best of our knowledge, there have been no three-dimensional (3D) boron carbonyl complexes observed to date in experiments, leaving an important question to be addressed in this area.

Based on the fact that neutral B₂₀ has a perfect double-ring (DR) tubular D_10d structure as its well-defined global minimum (GM),^20–22 B₂₀⁺ monocation possesses a slightly distorted DR tubular D_2d configuration as its deep-lying GM, which was the only stable isomer of the system observed in ion-mobility experiments.²³ While both of them can be considered as the embryos of single-walled boron nanotubes, we report herein the observation of a series of DR tubular boron carbonyl monocations B₂₀(CO)_n⁺ (n = 1–8) in gas phase under ambient conditions, as revealed by detailed chemisorption experiments and extensive first-principles theory calculations, presenting the largest boron carbonyl complexes observed to date, which mark the 2D-to-3D structural transition in boron carbonyl complexes. DR tubular D_2d B₂₀⁺ with twenty periphery boron atoms was found to be much more reactive to chemisorb the first CO than the previously reported planar π-aromatic C_2v B₁₃⁺, but much less reactive than both 2D C_s B₁₁⁺ (B₂@B₉⁺) and C_2v B₁₅⁺ (B₄@B₁₁⁺) with σ and π conflicting aromaticity. Detailed theoretical calculations and analyses reveal the chemisorption mechanisms and bonding patterns of the experimentally observed DR tubular B₂₀(CO)_n⁺ monocations and their neutral DR tubular counterparts B₂₀(CO)_n (n = 1–8), with both appearing to be aromatic in nature.

Methods

Experimental methods

A homemade reflection time-of-flight mass spectrometer (TOF-MS)^17,18,24,25 equipped with a laser ablation cluster source, a quadrupole mass filter (QMF),²⁶ and a linear ion trap (LIT)²⁷ reactor was employed in this work. Bare boron cluster monocations (B_n⁺) were generated by laser ablation of a rotating and translating ¹¹B target (99% enriched) under a 6 atm He carrier gas. A 532 nm pulsed laser (Nd³⁺:YAG second harmonic) with an energy of 3–5 mJ per pulse and a repetition rate of 10 Hz was used to ablate the boron target.

B₂₀⁺ monocations were mass-selected by the QMF and entered into the LIT, where they were confined and thermalized by collisions with a pulse of He gas and then interacted with a pulse of CO reactant gas. The cluster ions ejected from the LIT reactor were detected using TOF-MS. The pseudo-first-order rate constants (k₁) of the reaction between B₂₀⁺ and CO were determined using the following equation,

	(1)

in which I_R is the signal intensity of the reactant cluster ions after the reaction, I_T is the total ion intensity including contributions from the products, P_effective is the effective pressure of the reactant gas in the ion trap reactor, k_B is the Boltzmann constant, T is the temperature (∼300 K), and t_R is the reaction time. More details about the method to derive k₁ can be found in ref. 28.

Theoretical methods

Extensive density functional theory (DFT) calculations at the PBE0/6-311+G(d) level^29,30 were performed using the Gaussian 16 program.³¹ Previous CID experiments on BCAs B₁₃(CO)_n⁺ indicate that CO ligands in boron carbonyl complexes are coordinated molecularly to the peripheral atoms of the B_m⁺ cores, without being activated or dissociated.¹⁷ Using CO as molecular ligands to coordinate the experimentally observed DR tubular D_2d B₂₀⁺ core, which is the only stable isomer of the monocation existing in the gas phase,²³ the structures of the B₂₀(CO)_n⁺ (n = 1–8) boron carbonyl complexes, intermediates (IMs), and transition states (TSs) were extensively manually constructed and fully optimized in this work. The hybrid PBE0 functional, which has proven to be reliable for boron cluster calculations,^{17,18,32–34} was employed in this work. Vibrational frequency analyses confirmed that all IMs and TSs were true minima and transition states of the systems, respectively. Intrinsic reaction coordinate (IRC) calculations^35,36 were further conducted to verify that each TS with one imaginary vibrational frequency (as indicated in Fig. S10–S21) connects two appropriate IMs. Based on the fact that DR tubular D_2d B₂₀⁺ is the only isomer of the monocation observed in ion-mobility experiments²³ and CO ligands are molecularly coordinated to the B_m^+/0 cores in CID measurements,¹⁷ the optimized structures obtained herein by extensive manual structural constructions based on D_2d B₂₀⁺ are believed to be reliable, though no global structural optimization and direct IRPD spectroscopic comparison were performed at the current stage. PBE0-D3/6-311+G(d) structural optimizations with dispersion correction (Becke–Johnson damping D3 correction)³⁷ included were also performed to compare with the results obtained using PBE0/6-311+G(d). More accurate single-point DLPNO-CCSD(T)³⁸ calculations were performed on the most favorable chemisorption pathways of the experimentally observed B₂₀(CO)_n⁺ (n = 1–8) monocations using the ORCA program.³⁹ Adaptive natural density partitioning (AdNDP 2.0) bonding analyses^40,41 were performed on the concerned species. Iso-chemical shielding surfaces (ICSSs) were computed using the Multiwfn 3.8 program⁴² and visualized using the VMD 1.9.3 software.⁴³ Detailed anisotropies of the current-induced density (ACID) analyses were realized using the ACID 2.0 code⁴⁴ to further check the tubular aromaticity of the systems, with the ring current maps finally generated using POV-Ray 3.7 render.⁴⁵ In addition, energy decomposition analyses with natural orbitals for chemical valence (EDA-NOCV)^46–48 were performed using the ADF(2023) program package⁴⁹ at the PBE0/TZ2P level combined with the zeroth-order regular approximation (ZORA)⁵⁰ to elucidate the coordination bonding characteristics of B₂₀(CO)_n^+/0.

Results and discussion

Cluster reactivity measurements

The TOF mass spectra of the reactions of B₂₀⁺ with CO in Fig. 1A indicate that B₂₀⁺ can consecutively chemisorb up to eight CO ligands under ambient conditions to form the B₂₀(CO)_n⁺ (n = 1–8) complex series, presenting the highest coordination number of n = 8 in boron carbonyl complexes observed to date. Upon reacting B₂₀⁺ with 356 mPa CO for 2 ms, distinct signals of B₂₀CO⁺ and B₂₀(CO)₂⁺ corresponding to the first and second CO adsorptions were clearly observed, respectively (Fig. 1(a2)). When the CO pressure was increased to 893 mPa (Fig. 1(a3)), the product B₂₀(CO)₃⁺ emerged. Increasing the CO pressure to 1276 mPa (Fig. 1(a4)), the signal intensities of both B₂₀⁺ and B₂₀CO⁺ were significantly reduced, with a weak signal appearing at B₂₀(CO)₄⁺. To further investigate the chemisorption capacity of B₂₀⁺ toward CO, the reaction time was extended to 10 ms (Fig. 1(a5)). Under these conditions, the reactant signal of B₂₀⁺ and product signal of B₂₀CO⁺ disappeared completely, the B₂₀(CO)₂⁺ signal was obviously weakened, while the relative intensity of B₂₀(CO)₄⁺ increased effectively and new product signals of both B₂₀(CO)₅⁺ and B₂₀(CO)₆⁺ started to emerge. These results indicate that no inert isomers of B₂₀⁺ exist in our experiments. Further prolonging the reaction time to 20 ms (Fig. 1(a6)) and 30 ms (Fig. 1(a7)) led to a notable enhancement of the B₂₀(CO)₆⁺ signal and the ultimate emergence of both B₂₀(CO)₇⁺ and B₂₀(CO)₈⁺. No signal beyond, corresponding to the ninth CO adsorption product (B₂₀(CO)₉⁺), was detected in our experiments (Fig. 1(a7)). The experimental results observed above show that B₂₀⁺ can successively chemisorb up to eight CO molecules in the following chemisorption reaction:

B20+ + nCO → B20(CO)n+ (n = 1–8)	(2)

Based on detailed chemisorption measurements and the least-squares fitting procedure of eqn (1), the pseudo-first-order rate constants (k₁) for the reactions of B₂₀⁺ + nCO → B₂₀(CO)_n⁺ (n = 1–2) were estimated to be k₁ = (2.65 ± 0.53) × 10⁻¹¹ [ϕ = (1.91 ± 0.38)%] and (1.84 ± 0. 37) × 10⁻¹¹ [ϕ = (1.33 ± 0.27)%] cm³ molecule⁻¹ s⁻¹ for n = 1 and 2, respectively (Fig. 1B). For consecutive chemisorptions of B₂₀(CO)₂⁺ with more CO molecules to form B₂₀(CO)_n⁺ (n = 3–6), the k₁ values were estimated to be k₁ = (5.76 ± 1.15) × 10⁻¹¹, (1.10 ± 0.22) × 10⁻¹¹, (7.64 ± 1.53) × 10⁻¹³, and (1.95 ± 0.39) × 10⁻¹² cm³ molecule⁻¹ s⁻¹ for n = 3, 4, 5, and 6, with the corresponding reaction efficiencies of ϕ = (4.15 ± 0.83)%, (0.79 ± 0.16)%, (0.06 ± 0.01)%, and (0.14 ± 0.03)%, respectively (Fig. 1C). It is noticed that B₂₀(CO)₄⁺ possesses the highest mass intensity in Fig. 1(a4)–(a7) in the B₂₀(CO)_n⁺ (n = 2–8) series and exhibits the most abundant accumulative intensity in Fig. 1C in the B₂₀(CO)_n⁺ (n = 2–6) series. These kinetic results suggest that B₂₀(CO)₄⁺ possesses a unique structure, which is chemically more stable than its neighbours and relatively less reactive to chemisorb more CO molecules. Previous ion mobility measurements²³ indicate DR tubular D_2d B₂₀⁺ is the well-defined GM of the monocation and the only stable isomer observed in the gas phase. The k₁ value estimated here shows that the DR tubular D_2d B₂₀⁺ is about ten times more reactive to chemisorb the first CO than the typical planar π-aromatic C_2v B₁₃⁺ analogous to benzene,¹⁷ but about ten times less reactive than both quasi-planar C_s B₁₁⁺ and C_2v B₁₅⁺ with σ and π conflicting aromaticity,¹⁸ suggesting that relative reactivities of the concerned B_n⁺ towards the first CO are closely related to the geometries and aromaticities of the systems.

Chemisorption pathway analyses

The optimized lowest-lying isomers and corresponding most favourable chemisorption pathways of B₂₀(CO)_n⁺ (n = 1–8) monocations and B₂₀(CO)_n (n = 1–8) neutrals are depicted in Fig. 2A and C, respectively. Alternative low-lying isomers and minor chemisorption pathways starting from local minima, which are slightly less stable than the corresponding lowest-lying isomers on the most favourable chemisorption pathways, are collectively shown in Fig. S1–S6 and Fig. S8–S21, respectively. The CO molecular ligands appear to prefer to be chemisorbed along both the top and bottom B₁₀ rings of DR tubular B₂₀^+/0 at two opposite sides. As shown in Fig. 2A, the first CO is coordinated to the top B₁₀ ring of DR tubular B₂₀⁺ without an energy barrier, forming the most stable adduct C_s B₂₀(CO)⁺ (1A) with a chemisorption energy of 1.08 eV. Therefore, the observed B₂₀(CO)⁺ mass signal should correspond to C_s B₂₀(CO)⁺ (1A), which represents the most favourable configuration in both thermodynamics and dynamics. Similarly, the experimentally observed B₂₀(CO)₂⁺, B₂₀(CO)₃⁺, and B₂₀(CO)₄⁺ can be assigned to C₂ 2A, C₁ 3A, and C_s 4A (Fig. S8), which possess the most favourable chemisorption energies, respectively, via barrierless chemisorption processes. We notice that the experimentally observed C_s B₂₀(CO)₄⁺ (4A) can also be formed from the third lowest-lying isomer C₁ B₂₀(CO)₃⁺ (3C) in a barrierless minor chemisorption process (Fig. S9). Notably, with the largest chemisorption energy of 1.31 eV (Fig. 2A) and lowest reaction rate of k₁ = 7.64 × 10⁻¹³ cm³ molecule⁻¹ s⁻¹, the observed C_s B₂₀(CO)₄⁺ (4A) with four CO ligands symmetrically distributed at two opposite sites exhibits the highest mass intensity in the B₂₀(CO)_n⁺ (n = 1–8) series (Fig. 1(a4)–(a7)). The formation of B₂₀(CO)₅⁺ (5A) proceeds through the transition state TS1, which lies 0.07 eV lower than the entrance channel (Fig. 3A and Fig. S10). Further analysis reveals that the adsorption of the sixth CO molecule to form the most stable adduct C₂ B₂₀(CO)₆⁺ (6A) from C₁ B₂₀(CO)₅⁺ (5A) is also a barrierless process (Fig. 2A and Fig. S8). The subsequent chemisorption pathways C₂ B₂₀(CO)₆ (6A) + CO → C₁ B₂₀(CO)₇⁺ (7A) and C₁ B₂₀(CO)₇⁺ (7A) + CO → D₂ B₂₀(CO)₈⁺ (8A) involve the transition states TS2 and TS3, which lie 0.04 eV and 0.03 eV lower than the entrance channels (Fig. 2A and Fig. S11, S12), respectively, indicating that these processes are also barrierless, with the axially chiral D₂ B₂₀(CO)₈⁺ (8A) possessing the highest symmetry in the series with a symmetrical CO ligand distribution. However, as shown in Fig. S13, further calculations indicate that the chemisorption process of D₂ B₂₀(CO)₈⁺ (8A) + CO → C₁ B₂₀(CO)₉⁺ (9A) possesses a positive energy barrier of +0.15 eV, indicating that the formation of B₂₀(CO)₉⁺ is kinetically unfavourable under ambient conditions, consistent with the observation that no mass signal of B₂₀(CO)₉⁺ was detected in experiments (Fig. 1(a7)). As the most favourable chemisorption pathway, the D_2d B₂₀⁺ → C_s 1A → C₂ 2A → C₁ 3A → C_s 4A → C₁ 5A → C₂ 6A → C₁ 7A → D₂ 8A process in Fig. 2A possesses an overall exothermicity of 9.22 eV, enabling the consecutive chemisorptions of eight CO ligands in the B₂₀(CO)_n⁺ series (n = 1–8). As indicated in Fig. S7, the PBE0 and PBE0-D3 relative energies and energy barriers on the most favourable chemisorption pathway of the experimentally observed B₂₀(CO)_n⁺ monocations (n = 0–8) are well supported by single-point DLPNO-CCSD(T) calculations, the most accurate calculations performed in this work. Inspiringly, as clearly indicated in Fig. 2(B), the chemisorption energies (E_c) with respect to B₂₀⁺ + nCO → B₂₀(CO)_n⁺ exhibit an almost perfect linear relationship of E_c = 1.17n + 0.03 with the number (n) of CO ligands involved in the complexes, with the average chemisorption energy of E_c = 1.17 eV per CO, indicating that DR tubular B₂₀⁺ monocation consecutively chemisorbs the eight CO molecular ligands almost independently.

Further PBE0 calculations indicate that, starting from neutral DR tubular D_10d B₂₀, which possesses twenty equivalent peripheral boron atoms, the neutral C_s B₂₀(CO) (1a) can be formed by overcoming a marginal energy barrier of +0.02 eV, while C₂ B₂₀(CO)₂ (2a) and C₁ B₂₀(CO)₄ (4a) can be spontaneously generated via barrierless processes (Fig. 2C and Fig. S14, S15, S17). The sequential B₂₀(CO)_n can be achieved by overcoming the small energy barriers of +0.01, +0.04, +0.23, +0.04, and +0.05 eV for n = 3, 5, 6, 7, and 8, respectively (Fig. 2C and Fig. S16, S18–S21), with an overall exothermicity of 7.22 eV. The carbonylation of the typical aromatic neutral DR tubular D_10d B₂₀ turns out to be both thermodynamically and dynamically less favourable than that of the more electron-deficient DR tubular monocation D_2d B₂₀⁺. Interestingly, as shown in Fig. 2(D), the chemisorption energies with respect to B₂₀ + nCO → B₂₀(CO)_n also exhibit an almost perfect linear relationship E_c = 0.90n + 0.11 with the number of CO ligands involved in the systems, with an obviously lower average chemisorption energy of E_c = 0.90 eV per CO. Interestingly, as indicated in Fig. 2, with D3 dispersion corrections included, the PBE0-D3 approach produces similar optimized structures and relative energies with PBE0 for both the B₂₀(CO)_n⁺ and B₂₀(CO)_n (n = 1–8) series, with PBE0-D3 generating small negative energy barriers for all the concerned chemisorption processes, except B₂₀(CO)₅ (5a) + CO → B₂₀(CO)₆ (6a), which has a small positive energy barrier of +0.13 eV.

Bonding pattern and aromaticity analyses

To better understand the high stability of the B₂₀(CO)_n^+/0 (n = 0–8) series, detailed AdNDP bonding pattern analyses were performed, as shown in Fig. 3 and Fig. S22–S29. Fig. 3 clearly indicates that the perfect DR tubular neutral D_10d B₂₀ and experimentally observed DR tubular D_2d B₂₀⁺ possess very similar bonding patterns, which can be categorized into three subgroups. For simplicity, we first discuss the closed-shell neutral D_10d B₂₀ in Fig. 3A. The first subgroup in B₂₀ contains 20 2c-2e localized B–B σ-bonds on the two B₁₀ rings with the occupation numbers (ON) of ON = 1.79 |e|. These bonds can also be analysed as 20 3c-2e σ-bonds with slightly higher ON values (1.97 |e|), which are responsible for connecting the two adjacent B₁₀ rings, similar to the situation reported in DR tubular Co@B₁₆⁻.⁵¹ The second group consists of 5 20c-2e σ + σ bonds with ON = 2.00 |e|, which represent in-phase orbital overlaps between the two adjacent B₁₀ rings. These completely delocalized σ-bonds match the 4n + 2 Hückel rule (n = 2) and render σ-aromaticity to DR tubular D_10d B₂₀. The third group possesses 5 completely delocalized 20c-2e π + π bonds formed positively between the two B₁₀ rings with ON = 2.00 |e|, which match the 4n + 2 Hückel rule (n = 2) and make the system π-aromatic in nature. As clearly shown in Fig. 3B, the slightly distorted open-shell D_2d B₂₀⁺ possesses a bonding pattern very similar to that of neutral D_10d B₂₀, with the only difference occurring at the last delocalized 20c-2e π + π bond, which turns out to be singly occupied with ON = 1.00 |e|. For the corresponding boron carbonyl complex monocations, as demonstrated, Fig. 3C and D show the bonding patterns of C_s B₂₀(CO)₄⁺ and D₂ B₂₀(CO)₈⁺ on their DR tubular B₂₀ frameworks, respectively. Obviously, the 20 2c-2e localized σ-bonds formed on the two B₁₀ rings and 5 20c-2e σ + σ delocalized σ-bonds and 5 20c-2e π + π delocalized π-bonds formed positively over the DR tubular B₂₀ framework in D_2d B₂₀⁺ (Fig. 3B) have all been practically well inherited in both C_s B₂₀(CO)₄⁺ and D₂ B₂₀(CO)₈⁺, though with slightly lower ON values. Similar bonding patterns exist in the whole B₂₀(CO)_n⁺ and B₂₀(CO)_n species (n = 1–8), as depicted in Fig. S21–S28.

To comprehend the overall aromaticity of these DR tubular systems, Fig. 4 depicts the calculated ICSS surfaces of B₂₀(CO)_n⁺ and B₂₀(CO)_n (n = 2, 4, 6, 8) series, compared with that of the perfect DR D_10d B₂₀ on the top, which is known to possess typical tubular aromaticity.^20,52,53 Interestingly, both the B₂₀(CO)_n⁺ and B₂₀(CO)_n series exhibit similar ICSS surfaces with that of D_10d B₂₀, where the yellow areas with negative NICS-ZZ values inside the B₂₀ tube and within about 1.0 Å above the tube in the vertical direction belong to the chemical shielding regions, while the green regions with positive NICS-ZZ values like a belt around the B₂₀ tube in horizontal directions belong to the chemical de-shielding regions. More specifically, B₂₀(CO)_n⁺ and B₂₀(CO)_n (n = 2, 4, 6, 8) possess the negative nuclear independent chemical shift values of NICS = −19.75, −14.53, −7.46, and −5.36 and NICS = −26.06, −16.82, −9.57, and −4.94 ppm at the centres of the DR B₂₀ tubes, respectively. Similar ICSSs exist in other B₂₀(CO)_n⁺ and B₂₀(CO)_n (n = 1, 3, 5, 7) series (Fig. S30). These results indicate that, similar to DR tubular D_10d B₂₀, which has the largest negative NICS value of −39.71 ppm at the centre, both the distorted DR tubular B₂₀(CO)_n neutrals and B₂₀(CO)_n⁺ monocations (n = 1–8) exhibit tubular aromaticity, rendering extra-stability to help stabilize these DR tubular complexes. As indicated in Fig. S31, the calculated ring current maps of the slightly or severely distorted DR tubular B₂₀(CO)_n⁺ and B₂₀(CO)_n (n = 2, 4, 6, 8) using the ACID approach⁴⁴ further evidence the tubular aromaticity of the systems.

Effective σ-donations and weak π-back-donations

Detailed EDA-NOCV analyses in Fig. 5 and Tables S1, S2 reveal the corresponding deformation densities Δρ and shapes of the most important interacting orbitals of the pairwise orbital interactions depicted of C_s B₂₀(CO)⁺ and C_s B₂₀(CO), more specifically, with D_2d B₂₀⁺ and CO and D_10d B₂₀ and CO as reacting fragments, respectively. Obviously, one effective σ-donation from the HOMO of CO to the LUMO of B₂₀⁺ in radial direction, which contributes 65.37% to the overall orbital interaction (ΔE_orb = −105.2 kcal mol⁻¹) between B₂₀⁺ and CO and two weak π-back-donations perpendicular to each other from the HOMO and HOMO−9 of B₂₀⁺ to the two degenerate LUMOs of CO, which contribute 15.86% and 7.97% (Fig. 5A), respectively, coexist in the coordination interactions of C_s B₂₀(CO)⁺, with the dominant σ-donation bond being well reflected in the AdNDP 2c-2e B–C σ-bond in Fig. S21. As shown in the red → blue charge flow colour code, the doubly occupied σ-HOMO of CO serves as a lone-pair donor in the effective σ-donation interaction, while the two unoccupied degenerate antibonding π*-LUMOs of CO perpendicular to each other function as electron acceptors in the two weak π-back-donations. Similarly, one effective σ-donation, which contributes 58.76% to the overall orbital interactions and two weak π-back-donations, which contribute 21.07% and 8.71%, respectively, coexist in C_s B₂₀(CO), as clearly shown in Fig. 5B.

Similar coordination bonding patterns exist in the whole B₂₀(CO)_n^+/0 series between the CO ligands and the DR tubular B₂₀ core. The effective B₂₀ ← CO σ-donations and weak B₂₀ → CO π-back-donations in these boron carbonyl complexes appear to be similar to that of the TM–CO coordination interactions in classic TM carbonyl complexes, indicating that the peripheral boron atoms in these DR tubular boron carbonyl complexes can also be viewed as “honorable transition metals”,^4,5 similar to the situations observed in the previously reported 2D B₁₃(CO)_n⁺, B₁₁(CO)_n, and B₁₅(CO)_n⁺.^17,18

Conclusions

Extensive chemisorption experiments and first-principles theory calculations performed in this work indicate that previously experimentally observed gas-phase DR tubular D_2d B₂₀⁺ can chemisorb up to eight CO molecules consecutively under ambient conditions to form a series of DR tubular boron carbonyl monocations B₂₀(CO)_n⁺ (n = 1–8), unveiling the emergence of a 2D-to-3D transition in boron carbonyl complexes. Both DR tubular B₂₀(CO)_n⁺ and their neutral counterparts B₂₀(CO)_n appear to be tubularly aromatic in nature. Further joint experimental and theoretical investigations on larger boron carbonyl complexes B_m(CO)_n^+/−, which may have a 3D cage-like, bilayer and core–shell structures, are currently in progress. Carbonyl coordination is expected to be an effective approach to stabilize boron nanoclusters and low-dimensional nanomaterials to further enrich the chemistry of boron.

Conflicts of interest

There are no conflicts to declare.

Data availability

All the data are available online on the website of PCCP.

All the necessary low-lying isomers and their chemisorption pathways, bonding patterns, ICSS surfaces, ring currentmaps, and EDA-NOCV analyses provided. See DOI: https://doi.org/10.1039/d5cp01827g

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant No. 22003034, 22373061, and 92461303).

References

1. M. F. Zhou, L. Andrews and C. W. Bauschlicher, Jr., Chem. Rev., 2001, 101, 1931 CrossRef CAS .
2. Q. Q. Tian, X. Yin, R. J. Sun, X. F. Wu and Y. H. Li, Coord. Chem. Rev., 2023, 475, 214900 CrossRef CAS .
3. A. M. Ricks, Z. E. Reed and M. A. Duncan, J. Mol. Spectrosc., 2011, 266, 63 CrossRef CAS .
4. X. Wu, L. Zhao, J. Jin, S. Pan, W. Li, X. Jin, G. Wang, M. Zhou and G. Frenking, Science, 2018, 361, 912 CrossRef CAS .
5. S. K. Purkayastha, S. S. Rohman, P. Parameswaran and A. K. Guha, Phys. Chem. Chem. Phys., 2024, 26, 12573 RSC .
6. A. B. Burg and H. I. Schlesinger, J. Am. Chem. Soc., 1937, 59, 780 CrossRef CAS .
7. M. Zhou, N. Tsumori, L. Andrews and Q. A. Xu, J. Phys. Chem. A, 2003, 107, 2458 CrossRef CAS .
8. M. Zhou, N. Tsumori, Z. Li, K. Fan, L. Andrews and Q. Xu, J. Am. Chem. Soc., 2002, 124, 12936 CrossRef CAS .
9. M. Zhou, Z. X. Wang, P. V. R. Schleyer and Q. Xu, ChemPhysChem, 2003, 4, 763 CrossRef CAS PubMed .
10. T. R. Burkholder and L. Andrews, J. Phys. Chem., 1992, 96, 10195 CrossRef CAS .
11. M. Zhou, Q. Xu, Z. X. Wang and P. V. R. Schleyer, J. Am. Chem. Soc., 2002, 124, 14854 CrossRef CAS PubMed .
12. J. Y. Jin, G. J. Wang and M. F. Zhou, Chin. J. Chem. Phys., 2016, 29, 47 CrossRef CAS .
13. J. Jin, G. Wang and M. Zhou, J. Phys. Chem. A, 2018, 122, 2688 CrossRef CAS .
14. J. Jin, G. Wang, M. Zhou, D. M. Andrada, M. Hermann and G. Frenking, Angew. Chem., Int. Ed., 2016, 55, 2078 CrossRef CAS PubMed .
15. J. Jin and M. Zhou, Dalton Trans., 2018, 47, 17192 RSC .
16. Y. Zhao, T. Wang, C. Wang, Z. Zhang, H. Zheng, S. Jiang, W. Yan, H. Xie, G. Li, J. Yang, G. Wu, W. Zhang, D. Dai, X. Zheng, H. Fan, L. Jiang, X. Yang and M. Zhou, ChemPhysChem, 2022, 23, e202200060 CrossRef CAS .
17. R. N. Yuan, J. J. Chen, Q. Chen, Q. W. Zhang, H. Niu, R. Wei, Z. H. Wei, X. N. Li and S. D. Li, J. Am. Chem. Soc., 2024, 146, 31464 CrossRef CAS PubMed .
18. R. N. Yuan, Q. Chen, H. Niu, C. Y. Gao, X. N. Zhao, Y. B. Wu, S. G. He and S. D. Li, Phys. Chem. Chem. Phys., 2025, 27, 7279 RSC .
19. J. Wang, W. Fan, S. B. Cheng and J. Chen, J. Phys. Chem. A, 2024, 128, 7869 CrossRef CAS .
20. B. Kiran, S. Bulusu, H. J. Zhai, S. Yoo, X. C. Zeng and L. S. Wang, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 961 CrossRef CAS .
21. W. An, S. Bulusu, Y. Gao and X. C. Zeng, J. Chem. Phys., 2006, 124, 154310 CrossRef .
22. T. B. Tai, N. M. Tam and M. T. Nguyen, Chem. Phys. Lett., 2012, 530, 71 CrossRef CAS .
23. E. Oger, N. R. M. Crawford, R. Kelting and P. Weis, Angew. Chem., Int. Ed., 2007, 46, 8503 CrossRef CAS .
24. X. N. Wu, B. Xu, J. H. Meng and S. G. He, Int. J. Mass Spectrom., 2012, 310, 57 CrossRef CAS .
25. Z. Yuan, Y. X. Zhao, X. N. Li and S. G. He, Int. J. Mass Spectrom., 2013, 354, 105 CrossRef .
26. R. A. J. O’Hair, Chem. Commun., 2006, 1469 RSC .
27. L. D. Socaciu, J. Hagen, U. Heiz, T. M. Bernhardt, T. Leisner and L. Wöste, Chem. Phys. Lett., 2001, 340, 282 CrossRef CAS .
28. Z. Yuan, Z. Y. Li, Z. X. Zhou, Q. Y. Liu, Y. X. Zhao and S. G. He, J. Phys. Chem. C, 2014, 118, 14967 CrossRef CAS .
29. C. Adamo and V. Barone, J. Chem. Phys., 1999, 110, 6158 CrossRef CAS .
30. R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem. Phys., 1980, 72, 650 CrossRef CAS .
31. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, et al., Gaussian 16, Revision C.01, Gaussian, Inc., Wallingford, CT, USA, 2016 Search PubMed .
32. Q. Chen, W. L. Li, Y. F. Zhao, S. Y. Zhang, H. S. Hu, H. Bai, H. R. Li, W. J. Tian, H. G. Lu, H. J. Zhai, S. D. Li, J. Li and L. S. Wang, ACS Nano, 2015, 9, 754 CrossRef CAS .
33. H. J. Zhai, Y. F. Zhao, W. L. Li, Q. Chen, H. Bai, H. S. Hu, Z. A. Piazza, W. J. Tian, H. G. Lu, Y. B. Wu, Y. W. Mu, G. F. Wei, Z. P. Liu, J. Li, S. D. Li and L. S. Wang, Nat. Chem., 2014, 6, 727 CrossRef CAS .
34. J. Tian, X. N. Chen, S. D. Li, A. I. Boldyrev, J. Li and L. S. Wang, Chem. Soc. Rev., 2019, 48, 3550 RSC .
35. C. Gonzalez and H. B. Schlegel, J. Chem. Phys., 1989, 90, 2154 CrossRef CAS .
36. C. Gonzalez and H. B. Schlegel, J. Phys. Chem., 1990, 94, 5523 CrossRef CAS .
37. S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem., 2011, 32, 1456 CrossRef CAS PubMed .
38. Y. Guo, C. Riplinger, U. Becker, D. G. Liakos, Y. Minenkov, L. Cavallo and F. Neese, J. Chem. Phys., 2018, 148, 011101 CrossRef .
39. F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2022, 12, e1606 Search PubMed .
40. N. V. Tkachenko and A. I. Boldyrev, Phys. Chem. Chem. Phys., 2019, 21, 9590 RSC .
41. D. Y. Zubarev and A. I. Boldyrev, Phys. Chem. Chem. Phys., 2008, 10, 5207 RSC .
42. T. Lu and F. Chen, J. Comput. Chem., 2012, 33, 580 CrossRef CAS PubMed .
43. W. Humphrey, A. Dalke and K. Schulten, J. Mol. Graph., 1996, 14, 33 CrossRef CAS PubMed .
44. D. Geuenich, K. Hess, F. Köhler and R. Herges, Chem. Rev., 2005, 105, 3758–3772 CrossRef CAS .
45. Povray, Persistence of vision raytracer, POV-Ray 3.7, https://www.povray.org/.
46. T. Ziegler and A. Rauk, Theor. Chim. Acta, 1977, 46, 1 CrossRef CAS .
47. M. P. Mitoraj, A. Michalak and T. Ziegler, J. Chem. Theory Comput., 2009, 5, 962 CrossRef CAS .
48. M. Mitoraj and A. Michalak, Organometallics, 2007, 26, 6576 CrossRef CAS .
49. G. T. E. Velde, F. M. Bickelhaupt, E. J. Baerends, C. F. Guerra, S. J. A. Van Gisbergen, J. G. Snijders and T. Ziegler, J. Comput. Chem., 2001, 22, 931 CrossRef .
50. E. VanLenthe, E. J. Baerends and J. G. Snijders, J. Chem. Phys., 1993, 99, 4597 CrossRef CAS .
51. I. A. Popov, T. Jian, G. V. Lopez, A. I. Boldyrev and L. S. Wang, Nat. Commun., 2015, 6, 8654 CrossRef CAS PubMed .
52. H. J. Zhai, B. Kiran, J. Li and L. S. Wang, Nat. Mater., 2003, 2, 827 CrossRef CAS PubMed .
53. M. P. Johansson, J. Phys. Chem. C, 2009, 113, 524 CrossRef CAS .

This journal is © the Owner Societies 2025