Methyllithium and Its Oligomers as Lewis Bases for Hydrogen and Halogen Bonding

Abstract

When bonded to electropositive Li, the methyl C exhibits strong Lewis basicity. DFT calculations show a highly negative electrostatic potential on the C and a HOMO that closely resembles a C lone electron pair. LiMe and some of its oligomers are paired in the context of both halogen and hydrogen bonds with dihalogens and HX (X = F, I, Br, Cl). There is a wide variation of halogen bond strength from 1 to 20 kcal/mol, while the H‐bonds cover a narrower range between 4 and 10 kcal/mol. The electrostatic potential minimum and the binding of (LiMe)₂ are weakened relative to its monomer. Electrostatic effects account for slightly less than half of the total attractive energy in the halogen bonds and just over half for the H‐bonds. Enlarging the cluster to four and six Li atoms has a very mild weakening effect on the binding, reaching its asymptote for some 6–8 Li.

When bonded to electropositive lithium, the methyl carbon of LiMe exhibits significant Lewis basicity, so LiMe and some of its oligomers can form halogen bonds with X₂ or XF (X = Cl, Br, I) and hydrogen bonds with HX.

1. Introduction

There are many types of noncovalent interactions. Among them, perhaps the most fundamental and pervasive is the hydrogen bond [1] (HB) formalized by Pauling [2] in the 1930s. In addition, there are stacked structures with π–π interactions [3], phosphorus bond [4], chalcogen bond [5], and halogen bond (XB) [6]. The essence of these noncovalent interactions is a weak attraction, much less powerful than covalent bonds, which are associated with reversible chemical and biological processes. Their weak strength, reversibility, and diversity make them the cornerstone of dynamic biological processes [7], molecular recognition [8], and material assembly [9]. They play a crucial role in maintaining the structure and function of biological macromolecules [10], materials and nanotechnology [11], drug research and development, and chemical synthesis [12], together with environmental and energy applications [13].

A HB refers to the intermolecular interaction between an electron‐deficient H atom and a region of high electron density [14] on the partner unit, a directional force of medium strength. HBs play an important role in DNA replication [15], enzyme catalysis [16], protein folding [17], the design of nanomedicines [18], and the use of robust H‐bonded organic frameworks for pH‐responsive drug delivery [19]. When a halogen atom is covalently attached to a substituent, some electron density is drawn toward this bond and away from the region along the bond’s projection [20], yielding what is commonly termed a σ‐hole [21]. This region of positive electrostatic potential attracts an electron‐rich group to form a XB. These noncovalent bonds are involved in supramolecular self‐assembly [22], crystallography [23], the preparation of supramolecular low‐molecular‐weight gels [24], the design of molecules with high affinity and selectivity for proteins [25], and the application of XB principles in medicinal chemistry and chemical biology [26].

Lithium, an element central to the technology of the 21st century, has become the focus of much attention. Lithium‐based materials have broad application prospects due to their unique physical and chemical properties. They are used in organic lithium‐ion batteries [27], lithium–air batteries [28], synthetic drugs to improve the oxygen consumption rate of mitochondria [29], and the synthesis of lithium aluminosilicate (LAS) glass‐ceramics [30]. Organolithium compounds [31], an important branch of lithium chemistry, refer to a class of organometallic compounds containing carbon–lithium (C─Li) bonds. Due to their designable electronic structures and specific reaction sites, they are becoming key carriers for unlocking high‐performance materials [32] and also play a crucial role in reactions of organic [33] and organometallic chemistry [34]. Among them is lithium methide, the smallest organolithium compound. Although its application on an industrial scale is limited because it is extremely sensitive to oxygen and water and requires strict anhydrous and anaerobic conditions for experimental operations [35], it has an irreplaceable position in basic research and synthetic chemistry [36]. As one of the strongest organolithium bases and an excellent nucleophile, it is widely used in reactions such as deprotonation and addition. The aggregation state of lithium methide shows significant state dependence. It often exists as a tetramer in the solid state and mainly as a dimer in solution, and its aggregation state significantly affects its reaction activity [37].

Lithium atoms in lithium methide and its polymers act as Lewis acid centers to participate in the formation of lithium bonds [38]. As a second factor, its carbon atoms can serve as strong Lewis base centers, providing electron pairs to form noncovalent interactions with Lewis acids. The electron density distribution characteristics of methyl carbon, and its Lewis acidity and basicity, mainly depend on the electronegativity difference between the relevant atoms: when bonded to highly electronegative atoms (such as N, O, Cl), the electron density is drawn away from C, resulting in electron deficiency and Lewis acidity [39]. For example, this electron deficiency enables the methyl carbon to engage in tetrel bonding [40] as a Lewis acid. Conversely, when bonded to low‐electronegativity atoms (such as Li, B, Al), the electron density accumulates on the carbon atom, facilitating its activity as an electron donor [41] [42]. This potential reversal makes the methyl group a key switch for regulating intermolecular interactions. In the particular case of lithium methide, the bonding of C to the metallic Li makes the methyl C a strong nucleophile [43].

The goal of the current investigation is to thoroughly probe the electron‐donating ability of the methyl group when bound to the highly electropositive Li. The focus is placed on electron donation within the context of halogen bonding, where the methyl is paired with a wide range of electrophilic halogen centers. This analysis is extended to hydrogen bonding, another vehicle for electron donation by the methyl group. Of particular interest is the mode of interaction, given the fact that the tetracoordinated methyl group carbon has no formal lone electron pair, a typical requirement of the Lewis base in such bonds. Of further concern is the comparison of the lithium‐bonded methyl group with more typical electron donors that do in fact contain a robust lone pair. The scope of the project is extended to the dimer of LiMe which provides an interesting counterpoint, given the very different geometry, as explained in detail below, wherein the methyl C is symmetrically bound to the two Li centers, placing it in a pentacoordinate environment. The work finally progresses to larger systems containing up to six Li atoms so as to understand how the noncovalent bonding is modified and might be extrapolated to larger clusters.

2. Theoretical Methods

All calculations were performed using Gaussian 09 software [44]. The M06‐2X [45] density functional was selected for the calculations, combined with the triply polarized basis set def2‐TZVP [46]. All vibrational frequencies were positive, and each configuration was located at a local minimum on the potential energy surface. The Multiwfn software [47, 49] was used to analyze the molecular electrostatic potential (MEP) on the isosurface with an electron density of 0.001 atomic units and the average local ionization energy on the isosurface with 0.005 au. The intermolecular interaction energy (E _int) was calculated as the energy difference between the complex and the monomer sum. The counterpoise correction scheme proposed by Boys and Bernardi [50] was applied to correct the basis set superposition error (BSSE). Also calculated was the binding energy (E _b) which involves the energies of the subunits in their fully optimized geometries. The deformation energy DE represents the strain induced in the two monomers as they adapt to their structures within the dyad.

Multiwfn software provided the platform to implement atoms in molecules (AIM) [51] theory, viz., the electron density, its Laplacian, and energy density at the bond critical points (BCPs). Visual Molecular Dynamics [52] and Multiwfn softwares were used in combination for presentation. Weak interactions between complexes were identified and visualized through noncovalent interaction analysis (NCI) [53]. The Extended Transition State‐Natural Orbitals for Chemical Valence (ETS‐NOCV) [54] method was used to analyze the orbital interactions within the complex. Finally, the interaction energy was decomposed using the XEDA [55] software in combination with the GKS‐EDA [56] theory. The input file for XEDA was generated by the Mokit program, which analyzed various interactions in the experimental system and provided a reasonable explanation for the results. Its core advantage lies in the unified treatment of the energy contributions of covalent bonds, ionic bonds, and noncovalent bonds, and it is particularly suitable for the analysis of multielectron effects in organolithium systems.

This study systematically quantifies the orbital interactions and charge transfer (CT) components along the C···X/H direction in the complexes by combining the ETS‐NOCV [54] analysis with charge displacement (CD) [57] analysis based on a four‐component relativistic framework. The ETS‐NOCV‐CD curves were numerically evaluated using the PYCUBESCD program suite [58, 59]. Ciancaleoni et al. [60, 61] used CD to study various halogen bonded systems, confirming its feasibility.

3. Results

The next section comprises an analysis of the individual monomers, both Lewis bases and acids, with particular emphasis on their MEP. This segment is followed by a description of the complexes formed by pairing acids with bases. After a discussion of the energetics of these complexation reactions, the electronic structure is thoroughly analyzed, including the electron density topology and shifts induced by the complexation, and a decomposition of the total interaction energy.

3.1. Monomers

The bonding between the Lewis bases LiMe and (LiMe)₂ will depend, to some extent, on electron transfer from their occupied MOs, particularly their HOMO, to a vacant σ* orbital of the partner Lewis acid. The HOMO of these two bases is displayed in Figure 1a where they resemble a C lone pair in many respects, with some C─Li bonding character. The similarity between LiMe and (LiMe)₂ is also quite apparent with respect to this methyl region.

FIGURE 1.

(a) HOMO of LiMe and (LiMe)₂ where purple and green regions signify opposite sign of the wavefunction. (b) Molecular electrostatic potential (MEP) mapped onto the 0.001 au isodensity surface of monomers. Red and blue regions, respectively, represent positive and negative potential. All in kcal/mol.

The stability of HBs and XBs derives in large part from the directional Coulombic attraction between the positive electrostatic potential regions on the surface of Lewis acid and the electron‐rich regions of the base. The left side of Figure 1b contains the MEP maps of LiMe and (LiMe)₂ on their 0.001 au isodensity surface. As listed in Table 1, the minimum V _s,min of LiMe is −40.50 kcal/mol and that of (LiMe)₂ is −26.31 kcal/mol, both concentrated by the methyl carbon atoms. This highly negative MEP may be attributed to the electronegativity difference between lithium (χ= 0.98) and carbon (χ= 2.55) [35]. In the dimer structure of (LiMe)₂, a Li─C─Li─C rhombic structure is formed, and the carbon atom is bridged to lithium through two C─Li bonds. The charge is dispersed to the two lithium atoms, resulting in a weakened methyl σ‐lump.

TABLE 1.

Extrema on 0.001 au isodensity surface of X₂, XF, and HF monomers (V _s,max, kcal/mol).

	V s,max		V s,max(X)		V s,max(H)
F2	12.87	—	—	HF	68.72
Cl2	25.35	ClF	41.92	HCl	44.54
Br2	29.39	BrF	50.36	HBr	38.65
I2	32.14	IF	60.74	HI	28.71

The right side of Figure 1b displays the MEP of three Br‐containing Lewis acid molecules. (The MEP of all monomers is displayed in Figure S1.) There is a red positive σ‐hole near Br along the bond‐axis direction for X₂ and XF, while this positive region is associated with H for HX. The values of V _s,max on all of these acids are listed in Table 1, which range from 12.87 to 68.72 kcal/mol. These quantities vary with halogen atom size Cl < Br < I for X₂ and XF, and those of the latter are larger than those of the former. Note that V _s,max on the H changes in the opposite order for the HX series, commensurate with the diminishing electronegativity of X.

3.2. Geometries of Complexes

The pairing of each Lewis acid with LiMe and (LiMe)₂ leads to complexes whose structures are summarized in Figure 2. (Geometries of all such dyads are pictured in Figure S2.) In all cases, there is a close alignment between the negative MEP near C and the σ‐hole of the electron acceptor atom, whether X or H. The intermolecular distance is represented by R and r is equal to the internal bond length within the acid unit. As a measure of the “umbrella folding” of the methyl group, α is defined as the average of the three X···CH angles. The intermolecular distance R is contained in the first column of Table 2, followed by the stretch Δr imposed on the internal bond by the formation of the complex. Along with this elongation, the next column contains the shift of the internal vibrational stretching frequency to which it corresponds. The difference between R and r represents a simple measure of any transfer of the bridging atom across to the methyl, as a value of zero might be taken as half transfer.

FIGURE 2.

Optimized structures of complexes defining geometrical parameters.

TABLE 2.

Properties of optimized geometries, geometrical parameters defined in Figure 2. Distances in Ǻ, angles in degrees. Also listed are the frequency shift of X–R stretching (cm⁻¹).

	R	Δr	ΔνXR	R–r	α
LiMe···F2	2.775	0.019	−164.0	1.393	68.2
LiMe···Cl2	2.928	0.038	−89.4	0.903	69.7
LiMe···Br2	2.895	0.060	−69.7	0.549	71.6
LiMe···I2	3.004	0.063	−37.7	0.289	72.3
(LiMe)2···F2	2.865	0.012	−95.4	1.490	65.7
(LiMe)2···Cl2	3.035	0.024	−61.5	1.024	66.7
(LiMe)2···Br2	3.048	0.034	−41.8	0.728	67.7
(LiMe)2···I2	3.175	0.036	−21.4	0.487	68.0
LiMe···ClF	2.758	0.045	−150.0	1.099	71.3
LiMe···BrF	2.650	0.073	−186.9	0.829	75.6
LiMe···IF	2.744	0.071	−125.6	0.772	77.1
(LiMe)2···ClF	2.864	0.029	−98.8	1.221	68.0
(LiMe)2···BrF	2.858	0.039	−86.0	1.070	69.4
(LiMe)2···IF	2.937	0.041	−57.4	0.995	70.6
LiMe···HF	2.045	0.018	−401.5	1.107	70.0
LiMe···HCl	2.193	0.025	−350.8	0.891	69.4
LiMe···HBr	2.168	0.031	−375.6	0.716	69.7
LiMe···HI	2.204	0.033	−381.5	0.564	69.6
(LiMe)2···HF	2.100	0.014	−310.9	1.166	67.7
(LiMe)2···HCl	2.299	0.017	−230.3	1.005	66.9
(LiMe)2···HBr	2.320	0.018	−194.0	0.881	66.9
(LiMe)2···HI	2.408	0.016	−174.1	0.785	66.5

There are several trends in evidence in Table 2 that are consistent with one another. In the first place, the intermolecular distances are consistently shorter for LiMe than for (LiMe)₂. This pattern conforms to the more negative V _s,min for the former that would tend to pull the Lewis acid in closer. This same concept applies to the larger stretches in the Lewis acid associated with LiMe, and the greater redshifts in its stretching frequency. Second, as the central X atom is enlarged, from F to I, there is a progressively longer intermolecular separation R that is consistent with the larger radius of X (with a small variance involving Br in some instances; see below). Heavier X yields a progressively larger stretch of the internal bondlength r, indicating a stronger bonding with the Lewis base. On the other hand, the redshift magnitude diminishes with larger X, likely attributable to its heavier mass which reduces the vibrational frequency. Note also that R–r drops with larger X, indicating higher degree of halogen transfer. Another point of consistency is the larger measures of interaction (viz., shorter R, larger Δr and Δν) associated with XF than with X₂, commensurate with σ‐hole depth. The H‐bonds in the bottom section of Table 2 are somewhat different in that the bridging atom is the much lighter H in all cases. So for example, the frequency changes are far larger in magnitude. Nonetheless, the patterns fit the idea that the σ‐hole on H is attenuated as X grows larger, leading to weaker H‐bonding.

3.3. Complexation Energies

The energetics of the complexations are presented in Table 3. The interaction energy E _int follows many of the same trends exhibited by the geometries. This quantity is smaller in magnitude for the (LiMe)₂ as opposed to the simple LiMe monomer. The XF dihalides present the strongest bonding which complements their deepest σ‐holes, rising quickly from FF to IF. The HBs with HX become progressively weaker as the X atom is enlarged and becomes less electronegative. There is a tight connection between the interaction energy and the σ‐hole depth, with correlation coefficients exceeding 0.955 (Figure S3). Overall, E _int ranges between only −1.4 kcal/mol for (LiMe)₂···F₂ all the way up to −20.4 kcal/mol for LiMe···IF. The next column of Table 3 lists the binding energy (E _b), the energy that takes the fully optimized monomers to the complex, so includes within it the deformation energy DE required for the monomers to distort into the structures they adopt within the dyad. E _b follows the same trends as E _int. Note that the largest deformation energies occur when LiMe associates with BrF and IF, which have the largest XF stretching of 0.07 Å (Table 2).

TABLE 3.

Interaction energy (E _int), binding energy (E _b), and deformation energy (DE) of the complexes, all in kcal/mol.

	E int	E b	DE		E int	E b	DE
LiMe···F2	−1.99	−1.77	0.22	(LiMe)2···F2	−1.39	−1.30	0.09
LiMe···Cl2	−5.87	−5.28	0.59	(LiMe)2···Cl2	−4.20	−3.87	0.33
LiMe···Br2	−9.01	−7.58	1.43	(LiMe)2···Br2	−6.04	−5.38	0.66
LiMe···I2	−10.58	−9.00	1.58	(LiMe)2···I2	−7.04	−6.30	0.74
LiMe···ClF	−9.88	−8.47	1.41	(LiMe)2···ClF	−7.20	−6.38	0.82
LiMe···BrF	−16.46	−12.13	4.33	(LiMe)2···BrF	−10.60	−9.00	1.60
LiMe···IF	−20.42	−15.45	4.97	(LiMe)2···IF	−13.73	−11.51	2.22
LiMe···HF	−10.32	−9.77	0.55	(LiMe)2···HF	−8.13	−7.56	0.57
LiMe···HCl	−7.15	−6.73	0.42	(LiMe)2···HCl	−5.24	−4.92	0.32
LiMe···HBr	−6.89	−6.36	0.53	(LiMe)2···HBr	−4.83	−4.53	0.30
LiMe···HI	−5.71	−5.23	0.48	(LiMe)2···HI	−3.78	−3.57	0.21

3.4. Electronic Structures

The theory of AIM analyzes the topology of the electron density so as to identify bond paths and evaluate certain of their important properties. The AIM molecular diagrams of several of these complexes containing I are presented in Figure 3, with the remainder in Figure S4. Each contains a bond path between the methyl C and the relevant X or H atoms; the salient properties of the bond critical point are listed in Table 4. The first property of interest is the electron density at this point in the first column, followed in the next column by its Laplacian. It is widely accepted that the density represents a strong indicator of bond strength. The magnitudes of ρ, coupled with the positive values of ∇² ρ, are both consistent with the characterization of these bonds as largely noncovalent. This concept is confirmed by the small positive values of the total energy density H in the succeeding column. The same designation would be inferred from the signs of the three eigenvalues λ _n of the density gradient, two negative and one positive. Small ellipticity (Є) reflects bonding anisotropy‐symmetric systems, with near‐linear bonding.

FIGURE 3.

AIM diagram of complexes. The key path is represented by a yellow path, where the orange dots indicate the positions of the key critical points.

TABLE 4.

Electron densities (ρ), Laplacians (∇² ρ), total energy densities (H), Hessian eigenvalues (λ ₁, λ ₂, λ ₃), and ellipticity (Є) at the X···C and H···C BCPs in the complexes. All in au.

	ρ	∇2 ρ	H	λ 1	λ 2	λ 3	Є
LiMe···F2	0.0123	0.0522	0.0022	−0.0065	−0.0064	0.0652	0.0065
LiMe···Cl2	0.0166	0.0516	0.0013	−0.0079	−0.0079	0.0675	0.0078
LiMe···Br2	0.0215	0.0549	0.0003	−0.0109	−0.0109	0.0768	0.0108
LiMe···I2	0.0216	0.0482	−0.0004	−0.0101	−0.0101	0.0684	0.0100
(LiMe)2···F2	0.0103	0.0456	0.0021	−0.0052	−0.0048	0.0556	0.0052
(LiMe)2···Cl2	0.0137	0.0451	0.0015	−0.0593	−0.0553	0.0565	0.0562
(LiMe)2···Br2	0.0162	0.0463	0.0009	−0.0716	−0.0645	0.0599	0.0673
(LiMe)2···I2	0.0158	0.0411	0.0005	−0.0607	−0.0606	0.0532	0.0572
LiMe···ClF	0.0225	0.0655	0.0008	−0.0118	−0.0118	0.0892	0.0117
LiMe···BrF	0.0343	0.0679	−0.0027	−0.0211	−0.0211	0.1101	0.0207
LiMe···IF	0.0339	0.0581	−0.0039	−0.0189	−0.0189	0.0960	0.0185
(LiMe)2···ClF	0.0184	0.0583	0.0013	−0.0085	−0.0077	0.0746	0.0084
(LiMe)2···BrF	0.0223	0.0589	0.0001	−0.0104	−0.0096	0.0789	0.0103
(LiMe)2···IF	0.0231	0.0535	−0.0008	−0.0103	−0.0089	0.0727	0.0102
LiMe···HF	0.0252	0.0629	−0.0009	−0.0246	−0.0246	0.1122	0.0240
LiMe···HCl	0.0204	0.0522	0.0006	−0.0170	−0.0170	0.0863	0.0167
LiMe···HBr	0.0225	0.0528	0.0002	−0.0190	−0.0190	0.0908	0.0186
LiMe···HI	0.0218	0.0498	0.0003	−0.0175	−0.0175	0.0849	0.0172
(LiMe)2···HF	0.0227	0.0605	−0.0001	−0.0213	−0.0203	0.1021	0.0209
(LiMe)2···HCl	0.0168	0.0457	0.0011	−0.0129	−0.0123	0.0709	0.0127
(LiMe)2···HBr	0.0167	0.0437	0.0009	−0.0126	−0.0119	0.0683	0.0125
(LiMe)2···HI	0.0149	0.0377	0.0009	−0.0102	−0.0097	0.0577	0.0101

For the halogen bonded systems, ρ ranges from 0.0103 to 0.0343 au, consistent with weak interactions; ρ varies between 0.0149 and 0.0252 au, for the HBs which is in line with the typical range of HBs. The bond critical point densities and other parameters conform to the patterns noted above for other aspects of these intermolecular bonds. Specifically, LiMe engages in stronger bonding than does (LiMe)₂, XF > X₂, halogen bonding strengthens with larger X, while H‐bonding with HX follows the opposite trend.

Other insights into the electronic aspects of these complexes may be achieved by examining the electron density flow that occurs when the two subunits engage with one another. Diagrams of these density shifts are included in Figure 4 for several representative systems. The purple and green colors, respectively, indicate density accumulation and depletion as a result of the complexation process. Figure 4a encompasses the XB in LiMe···BrF which contains some of the typical features of a XB. There is an overall polarization of each subunit, for example the shift of density from left to right that further depletes charge from the σ‐hole region of the Br, further amplifying its positive charge. Density also shifts from left to right within the Lewis base, with density added to the σ‐lump to the right of the C atom, enhancing its negative MEP.

FIGURE 4.

Electron density shifts in (a) LiMe···BrF, (b) (LiMe)₂···BrF, and (c) LiMe···HBr that occur upon complexation. Purple and green colors designate gain and depletion of density, respectively, at the ±0.001 au contour.

The electron density shift pattern in Figure 4b for the (LiMe)₂ analog is quite similar. Notably, there is very little participation from the additional atoms within the base, with nearly all of the density changes occurring in the vicinity of the methyl group participating in the XB. There is a small but noticeable reduction in the size of the purple and green lobes relative to Figure 4a, which is commensurate with the weaker overall interaction. The shifts within the LiMe···HBr HB system in Figure 4c is quite similar to the two XBs.

In addition to the internal polarizations, there is also a net transfer of charge from the Lewis base to the acid. This transfer can be quantified by summing the natural charges on each of the subunits, which is listed in the total CT columns of Table 5. This quantity conforms to many of the patterns discussed above for other properties. CT is larger for LiMe than for (LiMe)₂ and rises along with the halogen atom size for the various XBs. The XF Lewis acids are also prone to larger CT than are the X₂ homodimers. The CTs for the HBs tend to be a bit smaller and are less sensitive to the nature of the XF halogen atom, although there is a trend toward larger CT for heavier X.

TABLE 5.

NOCV orbital energy (E _NOCV, kcal/mol), charge transfer value associated with NOCV pair (CT, e), and charge transferred between subunits (total CT, e).

	E NOCV	CT	Total CT		E NOCV	CT	Total CT
LiMe···F2	−3.78	0.035	0.033	(LiMe)2···F2	−2.45	0.022	0.021
LiMe···Cl2	−7.68	0.082	0.074	(LiMe)2···Cl2	−4.92	0.054	0.049
LiMe···Br2	−12.97	0.136	0.127	(LiMe)2···Br2	−7.36	0.084	0.078
LiMe···I2	−14.27	0.146	0.129	(LiMe)2···I2	−7.80	0.088	0.079
LiMe···ClF	−13.03	0.118	0.105	(LiMe)2···ClF	−8.79	0.085	0.072
LiMe···BrF	−25.67	0.214	0.190	(LiMe)2···BrF	−12.87	0.124	0.106
LiMe···IF	−32.91	0.261	0.189	(LiMe)2···IF	−16.06	0.150	0.115
LiMe···HF	−6.06	0.053	0.027	(LiMe)2···HF	−4.97	0.046	0.023
LiMe···HCl	−5.80	0.051	0.035	(LiMe)2···HCl	−3.95	0.039	0.026
LiMe···HBr	−7.36	0.063	0.048	(LiMe)2···HBr	−4.29	0.040	0.031
LiMe···HI	−8.24	0.068	0.055	(LiMe)2···HI	−3.87	0.036	0.031

The NOCV procedure permits a decomposition of the CT occurring within each complex into specific orbital pairs, revealing a clear electron transfer path. As shown in Figure 5, for F–X···LiMe and F–X···(LiMe)₂ complexes, the NOCV deformation density map (Δρ ₁) illustrates electron transfer from the σ(C─Li) bonding orbital (mainly from the carbon end) of LiMe to the σ* antibonding orbital of the X─Y bond in the halogen acceptor, σ(C─Li) → σ*(X─Y). When Y ≠ X, the σ* orbital is lower in energy. The figure shows a notable increase in the red region between carbon and X, and a smaller blue region near X with stronger interaction energy, both indicating CT from carbon to the Lewis acid. In the complexes formed by (LiMe)₂, the CT occurs from the lone‐pair electron orbital of C to the antibonding lone‐pair electron orbital of X, LP(C) → LP*(X).

FIGURE 5.

The charge displacement curve of F–X···LiMe/F–X···(LiMe)₂ complexes and the corresponding NOCV pair deformation density map (Δρ ₁): (i) F–Cl···LiMe, (ii) F–Br···LiMe, (iii) F–I···LiMe, (iv) F–Cl···(LiMe)₂, (v) F–Br···(LiMe)₂, and (vi) F–I···(LiMe)₂. The isodensity surface value of the deformation density map is ±0.001 a.u., The direction of charge flow is indicated as red→blue (red: electron depletion, blue: electron accumulation). Black dots indicate the positions of atomic nuclei on the z‐axis, and black vertical stripes mark the boundaries between fragments.

Table 5 lists the NOCV orbital energies and associated CT values, which exhibit a strong positive correlation, confirming that CT is an important driving force for bonding. In all systems, CT > 0, indicating electron flow from the carbon atom to the halogen‐containing molecule. CT values for XF are significantly higher than those for X₂, as the greater electronegativity difference enhances the electron‐accepting ability of the σ* orbital. Notably, a relatively high CT (>0.20 e) combined with H < 0 in AIM analysis suggests partial covalent character in LiMe···BrF/IF systems, where strong orbital mixing exceeds electrostatic attraction.

CD analysis based on the NOCV decomposition is depicted by the red curves in Figure 5 and represents the cumulative density change from C to the acceptor atom from left to right, involving the entire system. The charge decrease is concentrated near the C atom, while accumulation is concentrated at the acceptor atom end, and the CD reaches a peak near the boundary between the two fragments. The formation of XB and HB is mainly dominated by the net transfer wherein charge flows from the Lewis base to the Lewis acid. The amount of CT involving this orbital pair (CT in Table 5) increases with the rise of orbital energy. In the (LiMe)₂ dimer, CT decreases systematically. The deformation density map (Δρ ₁) clearly shows that the donor–acceptor channel is the main bonding component.

Regarding the hydrogen bonding systems, the NOCV diagram and CT value associated with NOCV pair (CT) in Figure 5 clearly indicate electron flow from the carbon atom of LiMe to the σ*(H–X) orbital of HX. Notably, the CT values of the heavy‐halogen systems HBr/HI are significantly higher than those of the light‐halogen systems HF/HCl. This is mainly because the large‐size orbitals of I and Br enhance the electron‐accepting ability of σ*. Similar to the XB system, there is a strong positive correlation between the orbital energy (E) and the CT value. On the other hand, it is only the XBs for which the NOCV energy is tightly correlated with the full interaction energy, as documented by the correlation coefficient of 0.955 in Figure 6a, as compared to essentially no correlation for the H‐bonds in Figure 6b.

FIGURE 6.

Correlation between NOCV orbital energies (E _NOCV, kcal/mol) and interaction energies (E _int, kcal/mol): (a) halogen bonding systems and (b) hydrogen bonding systems.

3.5. Energy Decomposition

Using GKS‐EDA, the total interaction energy (E ^total) can be decomposed into physically meaningful components: electrostatic energy (E ^ele), exchange‐repulsion energy (E ^ex+rep), polarization energy (E ^pol), and dispersion energy (E ^disp). These quantities are listed in Table 6, whose trends reveal certain properties of these interactions. Each quantity separately follows the same trends as the total interaction energy: A sharp rise is observed as the X center of X₂ or FX is enlarged although there is much less sensitivity, and even a reversal, for the HX series. Also, the values drop as LiMe is replaced by its dimer.

TABLE 6.

Electrostatic energy (E ^ele), exchange‐repulsion energy (E ^ex+rep), polarization energy (E ^pol), sum of dispersion and electron correlation energy (E ^disp/corr), and total interaction energy (E ^total) in kcal/mol. %E ^ele, %E ^pol, and %E ^disp/corr represent the percentage of each of these three terms relative to their sum.

	E ele	E ex+rep	E pol	E disp/corr	E total	%E ele	%E pol	%E disp/corr
LiMe···F2	−3.60	6.83	−0.24	−4.99	−2.00	40.8	2.7	56.5
LiMe···Cl2	−8.85	14.06	−4.23	−6.86	−5.88	44.4	21.2	34.4
LiMe···Br2	−14.35	22.18	−8.26	−8.59	−9.02	46.0	26.5	27.5
LiMe···I2	−15.56	24.48	−11.09	−8.40	−10.57	44.4	31.6	24.0
(LiMe)2···F2	−2.71	6.07	−0.09	−3.80	−0.53	41.1	1.4	57.6
(LiMe)2···Cl2	−6.53	10.53	−2.54	−5.66	−4.20	44.3	17.2	38.4
(LiMe)2···Br2	−9.54	14.74	−4.28	−6.96	−6.04	45.9	20.6	33.5
(LiMe)2···I2	−10.10	15.84	−5.8	−6.97	−7.03	44.2	25.4	30.5
LiMe···ClF	−14.67	21.26	−8.51	−7.96	−9.88	47.1	27.3	25.6
LiMe···BrF	−27.76	39.75	−18.86	−9.60	−16.47	49.4	33.5	17.1
LiMe···IF	−30.15	42.22	−24.32	−8.16	−20.41	48.1	38.8	13.0
(LiMe)2···ClF	−10.84	16.05	−5.46	−6.94	−7.19	46.6	23.5	29.9
(LiMe)2···BrF	−16.36	22.84	−9.29	−7.79	−10.6	48.9	27.8	23.3
(LiMe)2···IF	−19.00	25.99	−13.30	−7.42	−13.73	47.8	33.5	18.7
LiMe···HF	−13.12	11.63	−5.90	−2.92	−10.31	59.8	26.9	13.3
LiMe···HCl	−10.14	11.60	−5.37	−3.24	−7.15	54.1	28.6	17.3
LiMe···HBr	−10.49	13.83	−6.41	−3.82	−6.89	50.6	30.9	18.4
LiMe···HI	−9.24	14.36	−6.37	−4.46	−5.71	46.0	31.7	22.2
(LiMe)2···HF	−10.66	10.16	−4.67	−2.95	−8.12	58.3	25.5	16.1
(LiMe)2···HCl	−7.50	8.82	−3.60	−2.95	−5.23	53.4	25.6	21.0
(LiMe)2···HBr	−7.18	9.28	−3.74	−3.19	−4.83	50.9	26.5	22.6
(LiMe)2···HI	−5.71	8.42	−3.17	−3.33	−3.79	46.8	26.0	27.3

The last three columns of Table 6 present the percentage contributions to the total attractive sum. The electrostatic term constitutes roughly half of this total attraction. The balance between the polarization and dispersion‐correlation is much more flexible. The polarization energy rises along with the size of the X atom for the halogen‐bonding dyads, from as little as 1% up to 1/3. At the same time, this same progression from F to I drops the disp/corr contribution from more than half down to less that 20%, The H‐bonds behave differently in that polarization represents a nearly constant 25%–30%, and dispersion a bit less, particularly for HF and HCl. Regarding the LiMe versus its dimer, halogen bonding by the former is a bit more dependent on polarization and less upon dispersion than is (LiMe)₂.

Figures S5–S7 present correlation plots between the total interaction and E ^ele, E ^pol, E ^disp/corr in halogen and hydrogen bonded systems separately. The electrostatic and polarization energies correlate nicely with E ^total for the XBs, with correlation coefficients of 0.98, but the relationship deteriorates for disp/corr. It is only the E ^ele that correlates well with the total for the HBs, with R ² = 0.95.

4. Discussion

There have been some earlier studies concerning the ability of a methyl group to participate in the formation of various noncovalent bonds. For example, when a methyl group covalently attached to alkaline earth metals Mg and Be engages in a triel bond as an electron donor, its central C acquires a pentacoordinate configuration [62]. Such methyl groups have been confirmed to act as unconventional Lewis bases in chalcogen bonds as well [63]. MgMe₂ serves as an efficient electron donor to various p‐block Lewis acids and widely participates in noncovalent interactions of triel, tetrel, pnictogen, chalcogen, and XBs [64]. Notably, when the methyl groups in BeMe₂, MgMe₂, and MeMgCl form regium bonds with Cu, Ag, and Au of Group IB, the interaction is surprisingly strong. The interaction energy of AuCl···MgMe₂ reaches up to −72.3 kcal/mol, far exceeding the range of conventional noncovalent bonds and displaying partial covalent interaction [65]. As earlier noted [41], the carbon in LiMe lacks a formal lone pair but can nevertheless act as a Lewis base in noncovalent interactions via its C─Li σ orbital.

In a quantitative context, the electrostatic potentials evaluated here are in good agreement with earlier reported data. For instance, Politzer et al. [20] calculated the maximum electrostatic potentials of F₂, Cl₂, and Br₂ at the B3PW91/6‐31G(d, p) level as 13.8, 23.8, and 29.1 kcal/mol, respectively, closely matching our results of 12.9, 25.3, and 29.4 kcal/mol. The σ‐hole depth of halogen‐bond donors increases with halogen atomic radius, consistent with findings by Zheng et al. [66]. For hydrogen‐bond donors, as the halogen radius increases, the H─X bond lengthens, bond energy weakens, and the positive electrostatic potential at the proton end gradually decreases. The monomer LiMe exhibits strong bond polarity (V _s,min = −40.50 kcal/mol), which is significantly attenuated in the dimer (−26.31 kcal/mol), reflecting the substantial effect of dimerization on the electronic structure of methyllithium. In the dimer, two LiMe molecules interact via a Li─Li bond, leading to partial delocalization or shielding of the negative charge density at the carbon end (Me), thereby reducing the strong electronegativity characteristic of the monomer [67]. The average local ionization energies [68, 70] of methyllithium and its oligomers were examined, and the data are summarized in Figure S8. More negative electrostatic potential is associated with a reduced average local ionization energy, making it easier to extract electrons.

As discussed above, the dimerization of the LiMe unit leads to a slightly weakened (LiMe)₂ nucleophile. The V _s,min on the methyl C drops in magnitude from −40.5 to −26.3 kcal/mol, and the interaction energies with the various electrophiles are weakened accordingly. There is also a dramatic change of bonding pattern within the (LiMe)₂ subunit. Rather than the tetrahedral tetracoordinated C atom in LiMe, this carbon adopts a pentacoordinate geometry in the dimer, with two symmetric C─Li bonds that have elongated from 1.964 to 2.089 Å. Despite this rearrangement, the HOMO maintains its shape in the dimer, containing a pseudo‐lone pair pointing directly along the trisector of the CH₃ group. The MEP follows this same pattern in that its minimum is located in this same direction for both the monomer and dimer. Consequently, the preferred direction of approach by the incoming electrophile is unchanged by the dimerization of LiMe.

It was of interest to consider how further enlargement of this system might affect its halogen and hydrogen bonding ability. The geometries of Li₄Me₂ and Li₆Me₂ are displayed in Figure S9, where the r(C─Li) bondlengths are around 2.0 Å, quite similar to the 1.964 Å in LiMe and the 2.000–2.089 Å range in Li₂Me₂. The added Li atoms tend to reduce the magnitude of V _s,min at the methyl C, down to −21.0 and −21.5 kcal/mol for the two larger systems, respectively. The weakening of this σ‐lump is associated with a corresponding lessening of the interaction energy with three XF electrophiles, as documented in Table 7, with the molecular diagrams contained in Figure S10. In fact, just as V _s,min has reached its asymptote of −21 kcal/mol, so too has the interaction energy plateaued at −6.0, −8.8, and −11.4 kcal/mol for ClF, BrF, and IF, respectively. It may be surmised that the effects of the added Li atoms has been saturated at n = 8.

TABLE 7.

Interaction energy (E _int, kcal/mol) of complexes wherein the methyl C atom donates charge to the σ‐hole of XF.

	E int
Li2Me2···ClF	−7.20
Li2Me2···BrF	−10.60
Li2Me2···IF	−13.73
Li4Me2···ClF	−6.09
Li4Me2···BrF	−8.91
Li4Me2···IF	−11.47
Li6Me2···ClF	−6.03
Li6Me2···BrF	−8.83
Li6Me2···IF	−11.44

The linear relationship between the NOCV orbital interaction and the interaction energy affirms that in halogen bonded systems the occupation of the σ* orbital is a primary source of stabilization for the formation of the complex, especially for the BrF/IF systems. For hydrogen bonding systems, the trend of E _NOCV is different from that of E _int. By integrating the three sets of data of E _NOCV, GKS‐EDA, and E _int, the peculiarity of Br in LiMe···HBr can be attributed to the three‐body coupling of orbital activity–electrostatics–repulsion driven by high polarizability. High polarizability triggers significant orbital relaxation (increase in |E _NOCV|) and enhanced induced dipole (abnormal increase in E ^ele) under a strong electric field. However, it also leads to the simultaneous amplification of Pauli repulsion (E ^rep) and exchange energy (E ^ex). Since the growth rate of the latter exceeds that of the former, the total interaction energy (E _int) decreases. This mechanism explains that in noncovalent interactions of heavy atoms, a single electronic structure index (E _NOCV or E ^ele) is insufficient to accurately predict the binding strength. This finding is consistent with the contention of Varadwaj et al. [71] that the strength of noncovalent interactions cannot be predicted solely by the molecular electrostatic potential.

When the Li in LiMe acts as a Lewis acid center, it can attract the fluorine atom in the XF molecule to form an XF···LiMe dimer. The geometric structures are summarized in Figure S11. The interaction energy is between −7.17 and −21.05 kcal/mol, which is similar to the interaction energy range of LiMe···XF. Although the IF···LiMe is more stable, it does not contain the interaction mode of interest.

The total energy (E ^total) is consistent with the interaction energy (E _int), but GKS‐EDA reveals detailed contributions. The stability of all systems is dominated by electrostatic energy (E ^ele), confirming electrostatic complementarity between the σ‐hole and the negatively charged carbon region. The BrF and IF systems differ significantly from classical XBs, exhibiting extremely high polarization energy (E ^pol) due to strong charge separation and high CT, which distort the carbon‐end electron cloud and increase acceptor polarizability. With polarization proportions exceeding 30%, these systems transition beyond classical electrostatic interaction into electrostatic‐polarization synergy. The HF system shows the highest electrostatic proportion (59.8%), with low E ^pol reflecting limited electron cloud distortion a and a low CT value confirming a nonclassical CT mechanism. Energy decomposition analysis indicates that LiMe dimerization weakens both E ^ele and E ^pol, reducing bonding strength accordingly.

The inequivalent I···H distances are observed in (LiMe)₂···IF trimer, so that one might wonder about the presence of secondary interactions in addition to the C···X bonds of interest. However, the AIM diagrams of the various complexes in Figures 3 and S4 argue against the existence of any such secondary bonding, as well as the results of natural bond orbital (NBO) analysis; all indicate that there is no such secondary bond interaction.

These results of this study indicate that the research on noncovalent bonding in the methyllithium system has clear application prospects in the field of functional materials: the partial covalency of XBs can enhance the stability of the crystal framework, which is suitable for the design of porous adsorption materials; the dispersion compensation effect of dimers can tolerate the solid‐state environment, providing a new strategy for the development of lithium‐battery electrolyte additives. In the future, the practicality of these theoretical designs needs to be verified through crystal synthesis and device testing (such as battery cycling performance).

5. Conclusions

Methyllithium LiMe and its dimer (LiMe)₂ engage in fairly strong complexes with X₂, XF, and HX. The two former molecule types form XBs via a σ‐hole on the X center, while HX acts through a HB. The principal site of electron donation is the HOMO of the electrophile which might be described as a pseudo‐lone pair on the methyl C atom. The dimerization of LiMe to (LiMe)₂ attenuates the negative potential near the C and weakens these noncovalent bonds, both XB and HB. As per the usual trends of XBs, the σ‐hole on X and the bond strength are both strengthened as X is enlarged Cl < Br < I, and FX forms stronger bonds than does X₂. The H‐bonding of HX follows the opposite trend of HF > HCl > HBr > HI. XBs occur with a wide range of interaction energies from 2.0 up to 20.4 kcal/mol; the HB range is 5.7–10.3 kcal/mol. XBs in the upper limits of bond strength appear to border on partial covalency. Electrostatics accounts for nearly half of the total attractive energy of the XBs, but makes a slightly larger contribution to the HBs. Enlargement of the (LiMe)₂ Lewis base by the addition of more Li atoms has a mild weakening effect on the halogen and hydrogen bonding. This effect reaches its asymptotic limit at 6–8 lithiums.

Funding

This work was supported by Natural Science Foundation of Shandong Province (ZR2021MB123) and grant to SS by the US National Science Foundation (1954310).

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.