Tripodal Podand Ligand with a Superhalogen Nature as an E ﬀ ective Molecular Trap

: Tris(2-methoxyethyl) ﬂuoroborate anion (TMEFA), anovel tripodal ligand based on the BF 4 − superhalogen anion, is proposed and was investigated theoretically using ab initio MP2 (second-order Møller-Plesset perturbational method) and OVGF (outer valence Green function) methods. The studied molecule comprises three 2-methoxyethoxy groups (-O-CH 2 -CH 2 -O-CH 3 ) connected to a central boron atom, which results in the C 3 -symmetry of the compound. The resulting anion was stable against fragmentation processes and its vertical electron detachment energy was found to be 5.72 eV. Due to its equilibrium structure resembling that of classical tripodal podands, the [F-B(O-CH 2 -CH 2 -O-CH 3 ) 3 ] − anion is capable of binding metal cations using its three arms, and thus may form strongly bound ionic complexes such as [F-B(O-CH 2 -CH 2 -O-CH 3 ) 3 ] − / Li + and [F-B(O-CH 2 -CH 2 -O-CH 3 ) 3 ] − / Mg 2 + . The binding energies predicted for such compounds far exceed those of the similar neutral classical podand ligands, which likely makes the [F-B(O-CH 2 -CH 2 -O-CH 3 ) 3 ] − system a more e ﬀ ective molecular trap or steric shielding agent with respect to selected metal cations.

Classical superhalogen anions can also be useful as electronegative components of ionic liquids [39], whereas some high-symmetry compact structures may be utilized as steric shielding agents with respect to selected metal cations [40]. The ability to trap positively charged ions and thus protect them from contact with the chemical environment, and to introduce selected cations into various solutions are general features of certain complexes formed by crown ethers, cryptands, and podands [41][42][43][44][45]. In general, podands are acyclic supramolecular ligands and open-chain analogs of macrocyclic systems, with nonrigid structures containing heteroatoms embedded into their molecular "arms", which enables the complexation of various ions and neutral molecules [46,47]. Due to their rather flexible structure, podands have no preformed cavities, which in turn allows them to bind molecular guests of various sizes. On the other hand, their inherently open-ended structure results in a low selectivity of this process and lower stability of the complexes formed [48,49].
In this work, we demonstrated the possible use of a novel superhalogen anion (i.e., tris(2methoxyethyl) fluoroborate anion, which we have termed TMEFA) as an effective molecular trap with respect to selected metal ions. Taking into account both the podand-like shape of TMEFA and its excess negative charge, this superhalogen anion is expected to be capable of binding cations inside its three arms and thus forming strongly bound ionic complexes characterized by binding energy values significantly exceeding those of systems involving classical podands. We believe that our results will inspire further investigations concerning this new class of podands and their properties in the near future.

Computational Methods
The equilibrium molecular structures and the corresponding harmonic vibrational frequencies of the [F-B(O-CH 2 -CH 2 -O-CH 3 ) 3 ] − system (i.e., tris(2-methoxyethyl) fluoroborate anion), its ionic complexes with Li + and Mg 2+ cations, and the analogous compounds formed by three arbitrarily chosen conventional tripodal podands (in order to compare the binding strength with respect to these cations) were calculated using the second-order Møller-Plesset (MP2) perturbational method [50,51] together with the Dunning's correlation consistent basis set of double-zeta quality augmented with a standard set of diffuse functions (aug-cc-pVDZ) [52][53][54].
Using the obtained equilibrium geometries, we applied the CBS-QB3 complete basis set method of Petersson and coworkers [55][56][57] to calculate the binding energies (BEs) between TMEFA and Li + and Mg 2+ cations and for the complexes containing typical podands. It should be noted that such calculated binding energies by definition involve zero-point energy corrections.
The thermodynamic stability of the tris(2-methoxyethyl) fluoroborate anion was verified through analysis of the Gibbs free energy value of the most probable fragmentation route (i.e., F − loss) which was obtained using the CBS-QB3 free energies (at T = 298.15 K) estimated for the structures optimized at the MP2/aug-cc-pVDZ theoretical level.
The vertical electron energy (VDE) value of TMEFA was evaluated at the OVGF/6-311++G (3df, 2pd) level for the MP2/aug-cc-pVDZ global minimum structure. The outer valence Green function OVGF method (B approximation) [58][59][60][61][62][63] together with analogous 6-311++G (3df, 2pd) basis sets was successfully used for various superhalogen anions in the past, and provided very good agreement with experimentally measured VDEs [9,64]. We also verified that the pole-strength (PS) value obtained for the TMEFA system was sufficiently large (PS = 0.91) to justify the use of the OVGF approximation, because this method is valid only when the pole-strength values of the outer valence ionizations are larger than 0.80-0.85 [65].
The Merz-Singh-Kollman scheme [66] was used to calculate the partial atomic charges. All calculations were performed with the GAUSSIAN16 (Rev.C.01) program package [67].

Structure and Stability of TMEFA
The initial geometry of the tris(2-methoxyethyl) fluoroborate anion (TMEFA) was based on the tetrahedral BF 4 − superhalogen anion [68], in which three fluorine ligands were replaced with the 2-methoxyethoxy groups (i.e., -O-CH 2 -CH 2 -O-CH 3 chains). The geometrical optimization of this initial structure resulted in the C 3 -symmetry equilibrium geometry of the TMEFA closed-shell monoanion depicted in Figure 1. As can be seen, this geometry is typical for tripodal podand ligands [46], which contain three identical open-ended chains linked together by a central atom (usually nitrogen or boron).

Structure and Stability of TMEFA
The initial geometry of the tris(2-methoxyethyl) fluoroborate anion (TMEFA) was based on the tetrahedral BF4 − superhalogen anion [68], in which three fluorine ligands were replaced with the 2methoxyethoxy groups (i.e., -O-CH2-CH2-O-CH3 chains). The geometrical optimization of this initial structure resulted in the C3-symmetry equilibrium geometry of the TMEFA closed-shell monoanion depicted in Figure 1. As can be seen, this geometry is typical for tripodal podand ligands [46], which contain three identical open-ended chains linked together by a central atom (usually nitrogen or boron).  which indicated the quasi-tetrahedral structure of the FBO 3 fragment. Since the TMEFA system has an excess negative charge, it is expected that the six oxygen donor atoms form a negatively charged cavity inside the three chain substituents, which in turn should allow for effective cation complexation. The results of the Merz-Singh-Kollman analysis performed for TMEFA indicated that the partial atomic charges on each of the oxygen atoms connected with boron were equal to −0.4 au, while the three remaining oxygen atoms held a negative charge of −0.3 au each. It should be emphasized that the partial atomic charge localized on the fluorine atom was also negative (−0.5 au); hence, one might anticipate the formation of ionic complexes comprising TMEFA and a metal cation in which the latter is placed between the fluorine atom and one of the oxygen atoms (i.e., outside the cavity). For this reason, we verified that such isomeric complexes (i.e., containing the F···M + ···O fragment) indeed represented the energy minima; however, they were much higher in energy (by more than 25 kcal/mol, which rendered their formation rather impossible under standard conditions) than the structures with metal ions bound inside the cavity (described in the following section).
In order to verify the thermodynamic stability of TMEFA against decomposition, we considered the only reasonable fragmentation path, namely, F − loss. The Gibbs free energy value estimated for this reaction was equal to 49.1 kcal/mol which confirmed the thermodynamic stability of the anion investigated. This means that the Gibbs free energy estimated for the opposite reaction, was strongly negative, which indicates that F − could be spontaneously attached to tris(2-methoxyethyl) borate at room temperature. We also verified that this process could proceed with no kinetic barrier and thus may represent a potential route of TMEFA synthesis.
Since TMEFA is a negatively charged species, we also evaluated the excess electron-binding energy of this anion. The VDE of the [F-B(O-CH 2 -CH 2 -O-CH 3 ) 3 ] − system was calculated to be 5.72 eV, which confirmed its superhalogen nature; however, it should be mentioned that the vertical electronic stability of TMEFA was significantly lower than that of its BF 4 − precursor (VDE = 8.98 eV, as calculated at the same theory level [67]). Thus, our results were consistent with the previous investigations devoted to this subject which indicated that introducing various ligands into a conventional superhalogen anion reduces its VDE value [69]. Summing this section up, we concluded that the TMEFA anion adopts a C 3 -symmetrical equilibrium structure and represents an electronically and thermodynamically stable system that is likely capable of acting as tripodal ligand with respect to metal cations.

Complexes of TMEFA with Li + and Mg 2+ Ions
After describing the tris(2-methoxyethyl) fluoroborate anion, which has been identified as a podand ligand with a superhalogen nature, we now move on to the discussion of our results concerning its potential use as an effective molecular trap. As indicated in the introduction, such supramolecular systems are rather structurally flexible; thus, the cavity size predicted for isolated TMEFA does not necessarily correspond to that expected in complexes formed by TMEFA and metal cations. This anticipation is also supported by our earlier findings for another superhalogen anion, the structure of which resembles a molecular basket and captures guest atoms inside its cavity [39]. Hence, we decided to examine the complexation ability of the TMEFA system with respect to two arbitrarily selected metal ions with either +1 or +2 excess charge (i.e., Li + and Mg 2+ ).
The  Figure 2. It needs to be emphasized that we limit our discussion herein to the lowest-energy structures of TMEFA/Li + and TMEFA/Mg 2+ systems, as the relative energy values of other isomeric structures were found to be too large to be competitive with the global minima (see the preceding section).  Figure 2, due to the interactions between oxygen atoms and metal ions, the structures of the resulting compounds were more compact than that of isolated TMEFA ligand. The Li-O distances (2.06 Å and 2.14 Å) and the Mg-O separations (2.08 Å and 2.13 Å) indicated that in both cases, the TMEFA acted as hexadentate podand ligand (i.e., a ligand forming six bonds with a guest atom). The F-B, B-O, O-C, C-C, C-O, and O-C bond lengths were only slightly changed in comparison to the isolated TMEFA anion described in the preceding section. In particular, the B-F distances in TMEFA/Mg 2+ and TMEFA/Li + complexes were found to beshortened by 0.1 Å and 0.07 Å, respectively, while all remaining bond lengths changed by less than 0.04 Å (see Figures 1 and 2 for details).
In order to verify whether the tris(2-methoxyethyl) fluoroborate anion might act as a molecular trap or play a steric shielding agent role, we decided to compare the binding energies (BE) of the complexes formed by TMEFA ligand to the BE values characterizing compounds involving three arbitrarily chosen (yet representative) tripodal podands. Two of these systems contained nitrogen as an anchor atom and two types of chains, namely -O-CH2-CH2-O-CH3 (as in the TMEFA ligand) and -

The lowest energy isomers of the [F-B(O-CH 2 -CH 2 -O-CH 3 ) 3 ] − /Li + and [F-B(O-CH 2 -CH 2 -O-CH 3 ) 3 ]
− /Mg 2+ complexes corresponded to C 3 -symmetrical structures with the metal cations localized on the C 3 -axis inside the cavities formed by the three molecular arms. As can be seen in Figure 2, due to the interactions between oxygen atoms and metal ions, the structures of the resulting compounds were more compact than that of isolated TMEFA ligand. The Li-O distances (2.06 Å and 2.14 Å) and the Mg-O separations (2.08 Å and 2.13 Å) indicated that in both cases, the TMEFA acted as hexadentate podand ligand (i.e., a ligand forming six bonds with a guest atom). The F-B, B-O, O-C, C-C, C-O, and O-C bond lengths were only slightly changed in comparison to the isolated TMEFA anion described in the preceding section. In particular, the B-F distances in TMEFA/Mg 2+ and TMEFA/Li + complexes were found to beshortened by 0.1 Å and 0.07 Å, respectively, while all remaining bond lengths changed by less than 0.04 Å (see Figures 1 and 2 for details).
In order to verify whether the tris(2-methoxyethyl) fluoroborate anion might act as a molecular trap or play a steric shielding agent role, we decided to compare the binding energies (BE) of the complexes formed by TMEFA ligand to the BE values characterizing compounds involving three arbitrarily chosen (yet representative) tripodal podands. Two of these systems contained nitrogen as an anchor atom and two types of chains, namely -O-CH 2 -CH 2 -O-CH 3 (as in the TMEFA ligand) and -CH 2 -CH 2 -N=CH-CH 3 , whereas the third compound contained boron as the central atom and -CH 2 -CH 2 -N=CH-CH 3 groups. The schematic structures of these N(CH 2 -CH 2 -N=CH-CH 3   According to our findings, the equilibrium structures of the N(O-CH2-CH2-O-CH3)3 compound and its complexes with Li + and Mg 2+ exhibit C3-symmetry, while those involving -CH2-CH2-N=CH-CH3 chains correspond to C1-symmetry (see Table 1). It should be emphasized that the N(CH2-CH2-N=CH-CH3)3 and N(O-CH2-CH2-O-CH3)3 ligands are tetradentate, which means that they interact with Li + or Mg 2+ ion by forming four dative bonds (i.e., involving their four atoms acting as lone-pair donors) (see Figure 3). On the other hand, the third podand considered (i.e., B(CH2-CH2-N=CH-CH3)3) has been found to act as a tridentate ligand, utilizing its three atoms as lone-pair donors, which in turn leads to the formation of complexes containing three dative bonds (Figure 3)  According to our findings, the equilibrium structures of the N(O-CH 2 -CH 2 -O-CH 3 ) 3 compound and its complexes with Li + and Mg 2+ exhibit C 3 -symmetry, while those involving -CH 2 -CH 2 -N=CH-CH 3 chains correspond to C 1 -symmetry (see Table 1). It should be emphasized that the N(CH 2 -CH 2 -N=CH-CH 3 ) 3 and N(O-CH 2 -CH 2 -O-CH 3 ) 3 ligands are tetradentate, which means that they interact with Li + or Mg 2+ ion by forming four dative bonds (i.e., involving their four atoms acting as lone-pair donors) (see Figure 3). On the other hand, the third podand considered (i.e., B(CH 2 -CH 2 -N=CH-CH 3 ) 3 ) has been found to act as a tridentate ligand, utilizing its three atoms as lone-pair donors, which in turn leads to the formation of complexes containing three dative bonds (Figure 3). The Li-N and Mg-N distances in [N(CH 2 -CH 2 -N=CH-CH 3 ) 3  As far as the binding energy values showing the strengths of the cation-ligand interactions are concerned, we found them to be large for the complexes formed by TMEFA ligand (see Table 1). In particular, the BE value of 180.1 kcal/mol estimated for TMEFA/Li + was much larger than the BEs calculated for the complexes formed by the Li + ion and the other podands considered (BE = 81.5-104.4 kcal/mol), while the BE of 457.3 kcal/mol predicted for TMEFA/Mg 2+ greatly exceeded those evaluated for the complexes involving the N(CH 2 -CH 2 -N=CH-CH 3 ) 3 , N(O-CH 2 -CH 2 -O-CH 3 ) 3 , and B(CH 2 -CH 2 -N=CH-CH 3 ) 3 podands and Mg 2+ (BE = 246.5-297.6 kcal/mol). Clearly, the excess negative charge of the TMEFA ligand is one of the main reasons for its much stronger interactions with the metal cations (manifested by large BE values) in comparison to the complexes containing neutral podands. Therefore, taking into account (i) the electronic and thermodynamic stability of the tris(2-methoxyethyl) fluoroborate ligand, (ii) the predicted ease of its synthesis, and (iii) its high affinity to the metal ions, we may draw the conclusion that the TMEFA represents a promising candidate for an effective molecular trap and a steric shielding agent to protect selected cations from the chemical environment.

Conclusions
The electronic and thermodynamic stability of the tris(2-methoxyethyl) fluoroborate anion (TMEFA), as well as its potential application as a molecular trap for Li + and Mg 2+ cations, were investigated on the basis of ab initio quantum-chemical calculations carried out at the CBS-QB3//MP2/aug-cc-pVDZ and OVGF/6-311++G (3df, 2pd) theory levels. The results obtained led us to the following conclusions: 1.
The TMEFA system is a strongly electronically bound anionic compound with a superhalogen nature, thermodynamically stable against decomposition reactions, whose vertical electron detachment energy was found to be 5.72 eV.

2.
The equilibrium structure of TMEFA resembles those adopted by many well-known tripodal podand ligands, with a negatively charged cavity created by three open-ended 2-methoxyethoxy groups which allows for binding of selected metal ions (Li + and Mg 2+ ) inside the molecular "arms". 3. The

4.
Due to the anticipated ease of its synthesis and expected high affinity to various metal cations, TMEFA represents a promising novel podand ligand which might be utilized as an effective molecular trap and/or a steric shielding agent with respect to selected ions.