Unique reactivity of B in B[Ge9Y3]3 (Y = H, CH3, BO, CN): formation of a Lewis base

G. Naaresh Reddyac, Rakesh Paridaac, R. Inostroza-Riverab, Arindam Chakrabortyd, Puru Jena*e and Santanab Giri*a
aSchool of Applied Sciences and Humanities, Haldia Institute of Technology, Haldia, 721657, India. E-mail: santanab.giri@gmail.com
bFacultad de Ciencias de la Salud, Universidad Arturo Prat, Casilla 121, 1110939 Iquique, Chile
cDepartment of Chemistry, National Institute of Technology, Rourkela-769008, India
dFaculty of Science, Jatragachi Pranabananda High School, New Town, Kolkata, 700161, India
eDepartment of Physics, Virginia Commonwealth University, Richmond, VA, USA. E-mail: pjena@vcu.edu

Received 7th August 2019 , Accepted 19th August 2019

First published on 19th August 2019

Boron compounds usually exhibit Lewis acidity at the boron center due to the presence of vacant p-orbitals. We show that this chemistry can be altered by an appropriate choice of ligands to decorate the boron center. To elucidate this effect, we studied the interactions of boron with two classes of ligands, one based on penta-substituted phenyl species (C6X5, X = F, BO, CN) and the other based on Zintl-ion-based groups (Ge9Y3, Y = H, CH3, BO, CN). An in-depth analysis of the charges and Fukui function values at the local atomic sites of the substituted boron derivatives B(C6X5)3 and B[Ge9Y3]3 shows that the B-center in the former is electrophilic, while it is nucleophilic in the latter. The chemical stability of the B[Ge9Y3]3 species is shown to be due to the presence of strong 2c-2e bonds between the B and Ge centers. Thus, the general notion of the Lewis acid nature of a boron center depends upon the choice of the ligand.

The Lewis acid and Lewis base concepts, developed by G. N. Lewis in 1923,1 constitute a vital domain of acid–base chemistry and have been well-researched over the years. Elements with vacant or partially vacant available orbitals act as Lewis acid sites, while those with available electron pairs for donation act as Lewis bases. Boron and aluminum atoms, which have available vacant p orbitals, therefore serve as good Lewis acid candidates in their compounds. However, an interesting revelation about the electronic response of the boron center has been noted upon changes in the nature of its substituents. Boron atom in its substituted analogs is customarily no longer an electrophilic site; rather, it becomes nucleophilic, and hence is a Lewis base! This finding opened new vistas of seminal research towards deciphering the mechanistic courses of the reactivity tendencies of substituted boron complexes and their applications in organic syntheses. A number of studies2–14 on the synthesis and characterization of tri-coordinated boron complexes have revealed the experimental existence of nucleophilic boron centers which show Lewis basicity. Some literature studies feature substituted boron complexes with both electrophilic and nucleophilic B-sites.15,16 However, the modeling of Zintl-substituted trivalent boron complexes exhibiting reversal of reactivity upon changing the substituents has not been found to date in the literature.

At first, we studied tris(pentafluorophenyl)borane [B(C6F5)3],17–19 a well-known Lewis acid. Because of the presence of the electronegative F atoms bonded to the carbon centers of the C6 ring, pentafluorobenzene has a positive electron affinity (EA = 4.42 eV) and an electronegativity of 6.75 eV. This enables the pentafluorophenyl ligand to withdraw electron density from the B-center, leaving B with a charge of 0.884 |e|. As a result, the substituted boron derivative acts as a Lewis acid. The calculated ground state energy, natural bond orbital (NBO)20 charges on the B and C centers, and Wiberg bond indices of the [B(C6F5)3] molecule are given in the ESI (S1); the results further confirm the stability and electrophilic nature of the B-center. To further substantiate this finding, we considered other ligands, such as C6(CN)5 (EA = 4.25 eV) and C6(BO)5 (EA = 3.85 eV), which form [B(C6(CN)5)3] and [B(C6(BO)5)3], respectively. For all these cases, we found that the B center bears a positive charge. The optimized geometries and Cartesian coordinates of C6F5, C6(CN)5, C6(BO)5 and their B complexes with NBO charges are given in the ESI (S2 and S3).

With the above results as our benchmark, we designed a Zintl-substituted Lewis acid where the Zintl cluster acts as an electron-withdrawing ligand. From our earlier work on the Ge94− Zintl ion-based superhalogen Ge9R3 (R = CN, CF3, NO2),21 we first used the Ge9H3 cluster to model a Zintl-substituted Lewis acid complex, B[Ge9H3]3. The Ge9H3 species, which has 39 electrons, has a tendency to gain an electron to achieve the stable 40-electron configuration, a stability criterion satisfied by the Wade–Mingos rule.22 This expectation is consistent with the ionization energy (IE = 6.58 eV) and the positive electron affinity (EA = 2.87 eV) of the Ge9H3 species. The electronegativity of Ge9H3 is also 4.72 eV, which is larger than that of boron (4.31 eV). This electronegativity difference between the ligand and the central B atom is on par with our hypothesis that the more electronegative Ge9H3 will withdraw electron density from the B-center, thereby rendering B[Ge9H3]3 as a site for nucleophilic attack. From the structural similarity between B[Ge9H3]3 and B(C6F5)3, one would expect B[Ge9H3]3 to behave like a Lewis acid. However, an in-depth analysis of the NBO charges on the B and Ge centers of the B[Ge9H3]3 cluster, computed at different levels of theory (B3LYP,23,24 wB97XD,25 PBE26), told a different story (Fig. 1). The B-center bears negative charge and is nucleophilic. To examine if this result could be due to computational limitations, we calculated the NBO and Hirshfeld27 charges using different basis sets (SDD, def2-TZVPP28). The trends in the atomic charges on the B-center of the B[Ge9H3]3 species remained unaltered. The ground state optimized geometry, NBOs and Hirshfeld charges on the B center of the B[Ge9H3]3 complex calculated at different levels and basis sets are shown in Fig. 1. The optimized geometries and Cartesian coordinates of B[Ge9H3]3 and B[Ge9H3]3 are shown in the ESI (S4). All the calculations were performed using the Gaussian0929 suite of programs.

image file: c9cp04361f-f1.tif
Fig. 1 Ground state optimized geometry, NBOs and Hirshfeld charges on the B center of the B[Ge9H3]3 complex.

To confirm the dynamical stability of B[Ge9H3]3, we performed an Atom Centered Density Matrix Propagation (ADMP)30,31 simulation for 1.5 ps at a temperature of 300 K. The potential energy surface (Fig. 2(a)) showed no distortion of the B[Ge9H3]3 complex. We performed a full Quantum Theory of Atoms In Molecules (QTAIM)32,33 analysis on the B[Ge9H3]3 structure using AIMAll software34 to evaluate the bonding pattern. We recall that the critical points are points in real space where the density ρr(r) = 0 and that they can be classified according to their rank and signature; one bond critical point (BCP) r(3, −1) is located between the two nuclei and the ring critical point (RCP) r(3, +1). When ∇2ρ(r) is positive (negative), the electron density is locally depleted (concentrated).

image file: c9cp04361f-f2.tif
Fig. 2 (a) ADMP molecular dynamics simulation on [B(Ge9H3)3]. (b) Atoms in molecules (AIM) (green dots: BCP; red dots: RCP).

The electron density Laplacian at the mentioned critical points (CPs) provides a way to characterize the natures of the formed bonds. A negative value is consistent with a shared interaction (i.e. covalent interaction) and a positive value is associated with mainly electrostatic or ionic interaction. The AIM calculations (Fig. 2(b)) show that the three Ge9H3 ligands are covalently bonded with B. For the reactivity at the B center, the Fukui function35,36 for nucleophilic attack (f(r)+) is −0.28, while that for electrophilic attack is (f(r)) = 0.008. Thus, the dual descriptor, which is the difference between f(r)+ and f(r), yields a negative value. According to Morell et al.,37 if Δf(r) < 0, the site can hardly undergo nucleophilic attack; therefore, electrophilic attack is more favorable. Thus, the B center in B[Ge9H3]3 indeed acts as an electron donor and, hence, the B[Ge9H3]3 complex forms a Lewis base. Similar results emerge from studies of B[Ge9Y3]3 (Y = CH3, BO, CN). These results are summarized in the ESI (Table S1). From the literature, we found that for superatomic complexes,38 it is also possible to explain the hybridization using localized molecular orbitals. The resulting natural localized molecular orbitals (NLMOs) of the B[Ge9H3]3 complex involving the 2s-B and 2p-B atomic shells are shown in Fig. 3. The three degenerate NLMOs are oriented at about 120° from each other, leading to a trigonal planar arrangement. From the comparison of the three sp2 hybrid orbitals in the BH3 molecule, shown in Fig. 3, it is clear that the NLMOs derived from the 2s and 2p boron shells can be viewed as a set of B[Ge9H3]3 complex sp2 hybrids.

image file: c9cp04361f-f3.tif
Fig. 3 Differentiation of natural localized molecular orbitals (NLMOs) for the proposed B[Ge9H3]3 and BH3 molecules.

To gain more understanding of this alteration in the reactivity trends, we investigated the nature of the ligand and the role of the dual descriptors, which accurately describe the reactivity trends of atomic sites in a molecule. As noted before, the 39-electron Ge9H3 cluster is stabilized in the anionic form by becoming a 40-electron system, consistent with the Wade–Mingos rule, where the geometrical configuration of the Ge9H3 anion is nido. However, if Ge9H3 donates an electron, the cluster becomes a 38-electron system, attaining a closo configuration. Note that a closo configuration is more stable than a nido form. Thus, if the Ge9H3 ligand can attain the closo configuration by donating an electron to the B-center, B can possess an excess negative charge. This suggests that Ge9H3 can be stabilized in the cationic form rather than in the anionic form. We checked the dynamical stability of the [Ge9H3]+ cation cluster by performing an Atom Centered Density Matrix Propagation (ADMP) molecular dynamics simulation to generate the trajectory. The optimized geometries and Cartesian coordinates of [Ge9H3]+ cation and its corresponding trajectories are presented in the ESI (S5). Note that there is no marked deviation in energy; thus, [Ge9H3]+ cation is dynamically stable.

To further investigate the origin of the negative charge on the B-center, we calculated the ionization energy (IE) of the Ge9H3 ligand. The corresponding value (6.58 eV) is lower than the IE of boron (8.302 eV). This suggests that the Ge9H3 ligand is more likely to lose an electron than B; thus, the latter carries excess negative charge. This trend is opposite to what we observed with B(C6F5)3 as a ligand. Note that the IE of the C6F5 ligand (9.09 eV) is larger than that of B. Thus, whether B will gain or lose an electron will be determined by a competition between the ionization energies and electronegativities of B and the ligands. To substantiate this hypothesis, we studied other Zintl-based groups, such as Ge9(BO)3, Ge9(CN)3 and Ge9(CH3)3. The resulting tri-substituted boron complexes are shown in Fig. 4. The optimized geometries and Cartesian coordinates are shown in the ESI (S6). The calculated electronegativities and ionization energies are given in the ESI (Table S1).

image file: c9cp04361f-f4.tif
Fig. 4 Optimized geometries of the [Ge9Me3], [Ge9(BO)3], [Ge9(CN)3], B[Ge9Me3]3, B[Ge9(BO)3]3, and B[Ge9(CN)3]3 clusters with their B center NBO charges.

To obtain a detailed understanding of the reactivities of these compounds, we also performed dual descriptor analysis and examined the Fukui function of B[Ge9H3]3. This reactivity descriptor is calculated from the charge differences of a particular atom inside a molecule in its cation, neutral and anionic forms. Details of this calculation are given in the ESI.

To further confirm the Lewis base nature of B, we built a complex with a well-known Lewis acid, BF3, and compared the charge separation with that oft he BF3–NH3 complex, which is a known Lewis acid–base pair. The optimized geometries and NBO charges of BF3[B(Ge9H3)3] are given in Fig. 5. Adaptive natural density partitioning (AdNDP)39 analysis of BF3[B(Ge9H3)3] shows that the occupation number (ON) between the B–B bonds is 1.85 |e|. This supports the presence of a strong 2c-2e sigma bond between the B-atom of [B(Ge9H3)3] and the B-atom of the BF3 molecule, which equally share one electron (see Fig. 5). The optimized geometries and Cartesian coordinates of BF3, NH3, BF3–NH3 and BF3[B(Ge9H3)3] are shown in the ESI (S7 and S8).

image file: c9cp04361f-f5.tif
Fig. 5 Optimized geometries of the BF3[B(Ge9H3)3] and BF3–NH3 complexes along with their NBO charge separations between the Lewis acid–base centers and the AdNDP 2c-2e bond results of the BF3[B(Ge9H3)3] complex.

The binding energy of BF3[B(Ge9H3)3], namely 17.05 kcal mol−1, confirms the formation of the Lewis acid–base pair. The NBO charge separation also supports the claim that the [B(Ge9H3)3] complex is indeed a Lewis base, as NH3 is in the BF3–NH3 complex.

The calculation of the 2c-2e bond also suggests that a sigma bond exists between the two boron centers. These findings confirm that the boron complex [B(Ge9H3)3] behaves like a Lewis base and that the boron center, being electron rich, favors electrophilic attack.

We also optimized the BF3[B{Ge9(CH3)3}3] and BF3[B{Ge9(CN)3}3] complexes and calculated their NBO charges. From the calculated NBO charges, we found that in the B[Ge9(CH3)3]3 and B[Ge9(CN)3]3 complexes, the B center also behaves as a Lewis base. The optimized geometries and Cartesian coordinates of the BF3[B{Ge9(CH3)3}3] and BF3[B{Ge9(CN)3}3] complexes are shown in ESI, S8.

In this context, recent work by Hicks et al.40 is noteworthy. These authors established the existence of a nucleophilic aluminyl anion. Aluminum compounds, like boron, were hitherto considered to be electron-deficient molecules; they undergo a reversal in their reactivity trends when an anionic aluminium(I) nucleophile is stabilized by the dimethylxanthene ligand to produce the potassium aluminyl [K{Al(NON)}]2 (NON = 4,5-bis(2,6-diisopropylanilido)2,7-di-tert-butyl-9,9-dimethylxanthene).

Inspired by the above findings, we followed a similar strategy and coupled H+ and K+ ions with the anionic [B(Ge9H3)3] complex. The calculated Δf(r) values at the B-center of the [B(Ge9H3)3] complex were found to be negative, which is consistent with our findings. In fact, in [B(Ge9H3)3], the B center bears more negative charge than that in [B(Ge9H3)3], as revealed by NBO charge analysis. The optimized geometries, Cartesian coordinates and binding energies (positive values) of the complexes H[B(Ge9H3)3] and K[B(Ge9H3)3] are shown in the ESI (S9 and S10); the data indicate the stability of these molecules. Frontier molecular orbital analysis of these complexes (shown in the ESI, S11) clearly shows that the HOMO is mainly housed on the B-centers, thereby guaranteeing its nucleophilic nature.

In conclusion, our comprehensive study revealed a complete reversal in the reactivity trends of the B-center in the Zintl-based cluster B[Ge9Y3]3 (Y = H, CH3, BO, CN) compared to that of B(C6X5)3 (X = F, BO, CN). Detailed charge analyses on the various atomic sites using different basis sets and levels of theory showed that the B-center in B[Ge9Y3]3 (Y = CH3, BO, CN) is nucleophilic, while that in B(C6X5)3 (X = F, BO, CN) is electrophilic. This trend is further supported by analyzing the local reactivity indices through the Fukui function values. The existence of strong 2c-2e bonds between the B and Ge centers establishes the chemical stability of B[Ge9Y3]3. The sharp departure in the chemistry of B can be understood by comparing the relative strengths of the ionization energy and electronegativity of B with those of its ligands. For pentafluorophenyl-substituted ligands, the ionization energies are larger than that of B, while the reverse is the case with Zintl-substituted ligands. In addition, the electronegativities of the penta-substituted phenyl ligands are much larger than those of the Zintl-substituted ligands. These results are summarized in Table S1 (ESI). Thus, Zintl-substitution of a boron compound renders the B-center nucleophilic. Analogous behavior of an Al-center, reported recently, complements the conclusion reached in our study and demonstrates the power of ligands in manipulating the chemistry of compounds. We hope that this study will stimulate synthesis of these Zintl-based compounds where B acts as a base.

Conflicts of interest

There are no conflicts to declare.


This work is supported by the Department of Science and Technology INSPIRE award no. IFA14-CH-151, Government of India. RIR wishes to thank Fondo Interno de la VRIIP-UNAP, para el desarrollo de proyectos internos, VRIIP0112-18. P. J. acknowledges support by the U.S. DOE, Office of Basic Energy Sciences, Division of Material Sciences and Engineering under Award No. DE-FG02-96ER45579.


  1. G. N. Lewis, Valence and the Structure of Atoms and Molecules, 1923, p. 142 Search PubMed.
  2. J. Cid, H. Gulyas, J. J. Carbo and E. Fernandez, Chem. Soc. Rev., 2012, 41, 3558–3570 RSC.
  3. N. Dettenrieder, H. M. Dietrich, C. Schadle, C. Maichle-Mçssmer, K. W. Tçrnroos and R. Anwander, Angew. Chem., Int. Ed., 2012, 51, 4461–4465 CrossRef CAS.
  4. K. Nozaki, Y. Aramaki, M. Yamashita, S.-H. Ueng, M. Malacria, E. Lacüte and D. P. Curran, J. Am. Chem. Soc., 2010, 132, 11449–11451 CrossRef CAS.
  5. H. Braunschweig, Angew. Chem., Int. Ed., 2007, 46, 1946–1948 CrossRef CAS.
  6. T. B. Marder, Science, 2006, 314, 69–70 CrossRef CAS.
  7. A.-F. Pecharman, A. L. Colebatch, M. S. Hill, C. L. McMullin, M. F. Mahon and C. Weetman, Nat. Commun., 2017, 8, 15022 CrossRef CAS.
  8. Y. Segawa, M. Yamashita and K. Nozaki, Science, 2006, 314, 113–115 CrossRef CAS.
  9. L. Weber, M. Schneider and P. Lçnnecke, J. Chem. Soc., Dalton Trans., 2001, 3459–3464 RSC.
  10. R. Kinjo, B. Donnadieu, M. A. Celik, G. Frenking and G. Bertrand, Science, 2011, 333, 610–613 CrossRef CAS.
  11. D. Wu, Y. Li, R. Ganguly and R. Kinjo, Chem. Commun., 2017, 53, 12734–12737 RSC.
  12. H. Braunschweig, R. D. Dewhurst, L. Pentecost, K. Radacki, A. Vargas and Q. Ye, Angew. Chem., Int. Ed., 2016, 55, 436–440 CrossRef CAS.
  13. R. Böser, L. C. Haufe, M. Freytag, P. G. Jones, G. Hörner and R. Frank, Chem. Sci., 2017, 8, 6274–6280 RSC.
  14. Y. Wang and G. H. Rabinson, Science, 2011, 333, 530–531 CrossRef CAS.
  15. D. Wu, L. Kong, Y. Li, R. Ganguly and R. Kinjo, Nat. Commun., 2015, 6, 7340 CrossRef CAS.
  16. B. Wang, Y. Li, R. Ganguly, H. Hirao and R. Kinjo, Nat. Commun., 2016, 7, 11871 CrossRef CAS.
  17. W. E. Piers and T. Chivers, Chem. Soc. Rev., 1997, 26, 345–354 RSC.
  18. W. E. Piers, Adv. Organomet. Chem., 2004, 52, 1–76 CrossRef.
  19. G. Erker, Dalton Trans., 2005, 1883–1890 RSC.
  20. A. E. Read, L. A. Curtiss and F. Weinhold, Chem. Rev., 1988, 88, 899–926 CrossRef.
  21. G. N. Reddy, R. Parida and S. Giri, Chem. Commun., 2017, 53, 9942–9945 RSC.
  22. D. M. P. Mingos, Acc. Chem. Res., 1984, 17, 311–319 CrossRef CAS.
  23. A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652 CrossRef CAS.
  24. A. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785–789 CrossRef.
  25. J. D. Chai and M. Head-Gordon, Phys. Chem. Chem. Phys., 2008, 10, 6615–6620 RSC.
  26. P. John, K. B. Perdew and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868 CrossRef.
  27. F. L. Hirshfeld, Theor. Chim. Acta, 1977, 44, 129–138 CrossRef CAS.
  28. F. Weigend, Phys. Chem. Chem. Phys., 2006, 8, 1057 RSC.
  29. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci and G. A. Petersson, et al., Gaussian 09, revision B.01, Gaussian, Inc., Wallingford, CT, 2010 Search PubMed.
  30. H. B. Schlegel, J. Millam, S. S. Iyengar, G. A. Voth, A. D. Daniels, G. E. Scuseria and M. J. Frisch, J. Chem. Phys., 2001, 114, 9758–9763 CrossRef CAS.
  31. S. S. Iyengar, H. B. Schlegel, J. Millam, G. A. Voth, G. E. Scuseria and M. J. Frisch, J. Chem. Phys., 2001, 115, 10291–10302 CrossRef CAS.
  32. R. F. W. Bader, Atoms in Molecules: A Quantum Theory, Clarendon Press, Oxford, 1990 Search PubMed.
  33. R. F. W. Bader, Acc. Chem. Res., 1985, 18, 9–15 CrossRef CAS.
  34. A. Todd and T. K. Keith, Gristmill Software, Overland Park KS, USA, 2017, http://aim.tkgristmill.com.
  35. K. Fukui, Theory of Orientation and Stereoselection, Springer, Berlin, 1973, p. 134 Search PubMed.
  36. K. Fukui, Science, 1982, 218, 747–754 CrossRef CAS.
  37. C. Morell, A. Grand and A. Toro-Labbe, J. Phys. Chem. A, 2005, 109, 205–212 CrossRef CAS.
  38. A. Muñoz-Castro, Chem. Sci., 2014, 5, 4749–4754 RSC.
  39. D. Y. Zubarev and A. I. Boldyrev, Phys. Chem. Chem. Phys., 2008, 10, 5207–5217 RSC.
  40. J. Hicks, P. Vasko, J. M. Goicoechea and S. Aldridge, Nature, 2018, 557, 92–95 CrossRef CAS.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c9cp04361f

This journal is © the Owner Societies 2019