ABSTRACT Carbyne‐terminated, carbon‐flanked noble‐gas cations CH 2 NgCR + (Ng = He–Rn; R = CH, N, P, As) serve as a minimal platform for carbon‐only coordination of noble gases and for tuning asymmetric C–Ng–C bonding. Geometry optimizations and thermochemistry were performed at the MP2/def2‐TZVPP level. Most systems adopt a C 2v Y‐shaped structure, with CH 2 HeCCH + and CH 2 NeCCH + as C s exceptions. In contrast, no bound minima were located for CH 2 HeCN + and CH 2 NeCN + at the MP2 level, although Y‐shaped structures were obtained at the DFT level. The singlet is the lowest electronic state for most bound systems, although CH 2 NeCCH + and CH 2 ArCP + have lower‐energy triplet states, while CH 2 ArCCH + and CH 2 XeCP + show near‐degenerate spin states. A dissociation network considering different fragmentation limits and the CH 2 spin state was used to assess thermodynamic stability. The bound He and Ne complexes are destabilized by three‐body homolytic fragmentation, whereas dissociation into CH 2 (t) + NgCR + generally provides the most relevant dissociation limit for the heavier analogues. Several Ar and Kr derivatives remain close to thermoneutrality or are slightly unstable at 298 K, while the Xe and Rn systems are thermodynamically protected with respect to the dissociation pathways considered. NPA, Mayer bond‐order, QTAIM, and AdNDP analyses reveal pronounced asymmetry: the H 2 C–Ng contact is weak and becomes closed‐shell for Ne, whereas Ng–CR exhibits shared‐shell character and near‐single‐bond Mayer indices for Ar–Rn. EDA, performed at the ωB97X–D3/def2–TZVPP level, indicates predominantly electrostatic H 2 C–Ng stabilization and a larger orbital contribution to the Ng–CR interaction in the heavier noble‐gas systems.
ABSTRACT Incorporation of metallic atoms into fullerenes leads to the characterization of nonclassical cage landscapes, highlighting the markedly ellipsoidal Sc 2 @C 2v (4059)‐C 66 structure. In this work, we set out to explore Sc 2 @C 2v (4059)‐C 66 and its Stone‐Wales transformation isomer (Sc 2 @ C 2v ‐(3441)‐C 66 ) in terms of the resulting characteristics provided by the donation of six electrons from the metallic atoms into the π‐electron C 66 kernel in terms of induced magnetic field characteristics. The Sc 2 @C 66 species shows a long/short axis ratio surpassing that of C 70 , featuring two sets of three consecutively fused pentagons at the poles. From the resulting electron transfer from two encapsulated Sc atoms to the C 66 cage, resulting in a 72 π‐electron system that fulfills Hirsch's 2( N + 1) 2 spherical aromaticity rule ( N = 5). Our results support the formation of a spherical aromatic endohedral metallic fullerene (EMF) despite the elongated shape, which shows the absence of a direct Sc‐Sc bond. The Stone‐Wales transformation connecting both isomers modulates their local electronic and magnetic properties, having a lesser impact on the global aromatic properties. These findings highlight the persistence of spherical aromatic characteristics and adaptability in relation to structural rearrangements within an ellipsoidal EMF, as a suitable building block for applications in supramolecular chemistry and related fields.
ABSTRACT A hybrid LSTM–Transformer Variational Autoencoder (LTVAE) framework is introduced in this paper for learning compact and chemically meaningful latent representations from SMILES strings, providing a SMILES‐to‐latent‐to‐SMILES representation model trained on ZINC15 and evaluated on ZINC15 and an out‐of‐distribution dye dataset. In the proposed model, sequential and structural dependencies in molecular strings are encoded using a bidirectional LSTM, while a Transformer‐based decoder is employed to reconstruct valid molecular structures in the evaluated settings. When trained on a curated ZINC15 dataset, the framework achieves reasonable reconstruction performance, with 98.58% token‐level accuracy, 92.19% string‐level accuracy, a 99.30% chemical validity ratio, and an average Levenshtein distance of 0.29 using a 64‐dimensional latent space. Reconstruction under an out‐of‐distribution evaluation setting is evaluated on a structurally distinct dye dataset, where 95.56% token‐level accuracy and 85.30% validity are achieved. The learned latent space preserves trends aligned with both general molecular descriptors and dye‐relevant structural descriptors, particularly conjugation length and aromatic surface fraction, with Fsp providing complementary information on molecular saturation. Controlled perturbations in the latent space suggest smooth and chemically consistent structural variations, illustrating the continuity and interpretability of the learned latent manifold.
ABSTRACT Gallium‐modified ZSM‐5 (Ga‐ZSM‐5) zeolites are highly effective catalysts for the ethane‐to‐aromatics (ETA) process. In this work, the catalytic role of the Ga + , [GaO] + and [GaH 2 ] + species in Ga–ZSM‐5 is investigated using a cluster model and density functional theory (DFT) calculations, focusing on ethane dehydrogenation (EDH) and subsequent ethene aromatization steps. Anharmonic vibrational analysis reveals that the H–Ga–H asymmetric stretching vibration corresponds to the experimentally observed band. Moreover, the Interaction Region Indicator (IRI) analysis shows that the ethane‐zeolite interaction is dominated by van der Waals forces. Although ethane activation over [GaH 2 ] + remains energetically demanding during the EDH stage, with an activation free energy of 68.7 kcal mol −1 , the subsequent steps are comparatively more accessible, and the active site can be regenerated without the severe energetic penalty observed for the Ga + and [GaO] + pathways. Finally, C 6 ring closure is identified as the turnover frequency‐determining transition state species for the ETA process. These results provide detailed mechanistic insight into the role of [GaH 2 ] + in the ETA process and support the use of cluster models as reliable and efficient alternatives for investigating metal‐modified zeolite catalysis.
ABSTRACT Post‐transition‐state bifurcation, which plays a critical role in determining product selectivity, has gained significant attention in synthetic chemistry. A trispericyclic reaction has been reported, producing three different cycloadducts from a single transition state structure. Its mechanisms have been examined through elementary reaction processes and dynamical trajectory simulations. However, because the relevant static reaction paths and dynamical trajectories are embedded in a high‐dimensional configurational space, their mutual relationships and the potential‐energy features governing product branching are difficult to interpret without choosing a limited set of reaction coordinates. In this study, we use the Reaction Space Projector (ReSPer) method, which mathematically constructs a reduced‐dimensional chemical reaction space from a set of reference structures to uncover the dynamical branching mechanisms of the trispericyclic reaction. The two‐dimensional reaction space reveals that the reaction path network in the configurational space forms an arrowhead‐shaped landscape, differing from the traditionally recognized fork‐shaped landscape. Projecting dynamical trajectories onto the two‐dimensional reaction space shows that dynamic branching processes occur near the shoulder in the energy profile along the main reaction path from the transition state to the major cycloadduct, where the potential valley orthogonal to the reaction direction flattens out. These results demonstrate that ReSPer provides a useful reduced‐dimensional representation for relating static reaction‐path networks to dynamical branching trajectories, and may offer new descriptors for future branching‐ratio prediction in pericyclic reactions.
