Balancing Donor‐Acceptor and Dispersion Effects in Heavy Main Group Element π Interactions: Effect of Substituents on the Pnictogen⋅⋅⋅π Arene Interaction

Abstract High‐level ab initio calculations using the DLPNO‐CCSD(T) method in conjunction with the local energy decomposition (LED) were performed to investigate the nature of the intermolecular interaction in bismuth trichloride adducts with π arene systems. Special emphasis was put on the effect of substituents in the aromatic ring. For this purpose, benzene derivatives with one or three substituents (R=NO2, CF3, OCHO, OH, and NH2) were chosen and their influence on donor‐acceptor interaction as well as on the overall interaction strength was examined. Local energy decomposition was performed to gain deeper insight into the composition of the interaction. Additionally, the study was extended to the intermolecular adducts of arsenic and antimony trichloride with benzene derivatives having one substituent (R=NO2 and NH2) in order to rationalize trends in the periodic table. The analysis of natural charges and frontier molecular orbitals shows that donor‐acceptor interactions are of π→σ* type and that their strength correlates with charge transfer and orbital energy differences. An analysis of different bonding motifs (Bi⋅⋅⋅π arene, Bi⋅⋅⋅R, and Cl⋅⋅⋅π arene) shows that if dispersion and donor‐acceptor interaction coincide as the donor highest occupied molecular orbital (HOMO) of the arene is delocalized over the π system, the M⋅⋅⋅π arene motif is preferred. If the donor HOMO is localized on the substituent, R⋅⋅⋅π arene bonding motifs are preferred. The Cl⋅⋅⋅π arene bonding motif is the least favorable with the lowest overall interaction energy.


Table of Contents
Figure S1. Potential energy curves (in kJ mol -1 ) for idealized BiCl3 adducts with benzene derivatives (see Scheme 1 for details) a) with one substituent, b) with three substituents calculated at the PBE-D3/def2-QZVP level of theory. .  Table S15. 13 C NMR chemical shifts of the free monosubstituted arenes (C6H5R) and mixtures with BiCl3 in a 1:1 and 1:7 molar ratio, measured in CD3NO2 solution at ambient temperature.  Table S17. 13 C NMR chemical shifts of the free trisubstituted arenes (C6H3R3-1,3,5) and mixtures with BiCl3 and in a 1:1 and 1:7 molar ratio, measured in CD3NO2 solution at ambient temperature. The difference between the chemical shift of the BiCl3 adduct and the chemical shift of the free trisubstituted arene. . Potential energy curves (in kJ mol -1 ) for idealized BiCl3 adducts with benzene derivatives (see Scheme 1 for details) a) with one substituent, b) with three substituents calculated at the PBE-D3/def2-QZVP level of theory. Figure S2. Potential energy curves (in kJ mol -1 ) for the interaction potentials of Bi(CH3)3 adduct with selected benzene derivatives calculated at the PBE-D3/def2-QZVP level of theory. 143 Δq=q(adduct)-q(free). q -partial charge on a specific atom(s), Δq -difference between partial charge of an atom in BiCl3 adduct and in an unbound BiCl3 molecule.  Figure S6. Dispersion energy plots for equilibrium structures of Bi···π arene adducts computed at the DLPNO-CCSD(T)/cc-pVQZ (cc-pwCVQZ-PP for bismuth) level of theory with tightPNO settings.   Scheme S13. Numbering of the atoms in NMR calculations. Table S15. 13 C NMR chemical shifts of the free monosubstituted arenes (C6H5R) and mixtures with BiCl3 in a 1:1 and 1:7 molar ratio, measured in CD3NO2 solution at ambient temperature. δ-C 6 H 5 R 13 C-NMR (CD 3 NO 2 ) δ-C ipso δ-C ortho δ-C meta δ-C para δ-R  Table S16. The difference between the chemical shift of the arene in the mixture with BiCl3 and the chemical shift of the free monosubstituted arene.
13 C NMR studies in solution were carried out in order to investigated the interaction between dispersion energy donors i. e. BiCl3 and a series of monosubstituted arene ligands in a 1:1 and 7:1 molar ratio, in CD3NO2 solution. The experimental results of the solution 13 C NMR spectroscopy show changes for all analyzed systems and in almost all cases a high-field shift is observed for the chemical shifts of the arene in the mixture with BiCl3 compared to the respective free monosubstituted arene. Larger chemical shifts are observed as the content of BiCl3 is increased. In contrast to the ipso, ortho and meta carbons the para carbon shows only minor changes of the chemical shift and for about half of the analyzed systems a low-field shift is observed. In the case of the assummed benzaldehyde·BiCl3 adduct larger shifts were observed for the C atoms, but this most likely is due to the fact that the benzaldehyde preferentially coordinates over the O atom and not to the π system, implying that oxygen is a better donor than the π system. This is in accordance with the theoretical calculations as discussed and given in Table 6. The chemical shifts observed from the experiment are significantly smaller than those obtained from the theoretical calculations and a trend with regard to the substituents as found in the theoretical part, does not become obvious. Most likely, this is because we have to consider certain equilibria and interference with the polar solvent nitromethane (see Figure S4).

C-NMR (CD 3 NO 2 ) Δδ-C ipso Δδ-C ortho Δδ-C meta Δδ-C para
Δδ-R -1.20 0.76 -0.07 1.09 3.38 Table S17. 13 C NMR chemical shifts of the free trisubstituted arenes (C6H3R3-1,3,5) and mixtures with BiCl3 and in a 1:1 and 1:7 molar ratio, measured in CD3NO2 solution at ambient temperature. The difference between the chemical shift of the BiCl3 adduct and the chemical shift of the free trisubstituted arene.  Figure S19. Nitrobenzene molecule surrounded by explicit nitromethane molecules optimized at the PBE-D3/def2-SVP level of theory. Figure S20. Possible motifs of interaction between solvent and BiCl3 molecule as calculated at the PBE-D3/def2-QZVP level of theory. The interaction energy for structure A amounts to -41 kJ mol -1 while the interaction energy for structure B is -46 kJ mol -1 .
In order to assess the influence of the solvation on the computed NMR spectra we have chosen nitrobenzene and nitrobenzene adduct with the BiCl3 molecule and used the CPCM solvation model with acetonitrile as a solvent due to the fact that its dielectric constant (ε=36.6) is similar to the one of nitromethane. Additionally, we examined the influence of explicit solvation on the 13 C Δδ values for free nitrobenzene. Therefore, we optimized nitrobenzene surrounded by eleven nitromethane molecules as a first solvation shell at the PBE-D3/def2-SVP level of theory. The obtained 13 C δ values were compared to the 13 C NMR chemical shifts computed for nitrobenzene in gas phase. The structure of the system is shown in Figure S5 and the results both for CPCM as well as for explicit solvation are shown in Table S12. The results were compared to the 13 C Δδ values from the gas phase calculations. The CPCM model gives similar trend as the gas phase calculations for predicting the carbon chemical shifts. An estimate of the solvent effects using explicit molecules yields similar trends but also shows that for a quantitative agreement long MD simulations with snapshot NMR calculations would be necessary, which goes beyond the scope of the current investigation.

Experimental details of NMR measurements
1 H and 13 C{ 1 H} NMR spectra were recorded at ambient temperature in CD 3 NO 2 with an Avance III 500 spectrometer (Bruker) at 500.30 and 125.81 MHz, respectively, and are referenced internally to the deuterated solvent relative to Si(CH 3 ) 4 (δ = 0.00 ppm). The NMR spectra were processed using the software MestReNova (version 11.0.0-17609 1 / version 11.0.4-18998 2 ). Refrences: