
Web Release Date: January 20,
Myriad Planar Hexacoordinate Carbon Molecules Inviting Synthesis




and
Department of Chemistry and Center for Computational Chemistry, University of Georgia, Athens, Georgia 30602, and Department of Chemistry, School of Science, Beijing Institute of Technology, Beijing 100081, People's Republic of China
Received October 11, 2006
Abstract:
The prototypal hexacoordinate carbon (phC) molecule, CB62- (D6h), can be annulated by inserting arenes, olefins, or other one or two atom bridging groups into a perimeter B-B bond. Two single-atom units also can bridge opposite CB6 edges. These strategies allow the design of limitless phC derivatives based on CB6 as a building block unit. Illustrative examples, shown to be viable computationally, offer possibilities for experimental realization. These phC's involve multicenter deltahedral
bonding, common for boron but unusual for carbon, in the ring planes, as well as
delocalization.
Molecules with planar hexacoordinate carbons (phC) are exciting
prospects. They violate both the usual maximum tetracoordination
of carbon and its proclivity for three-dimensional bonding. Two
independent research groups discovered phC examples computationally in 2000. Exner and Schleyer's CB62- (Figure 1, 1) and the
isoelectronic C3B4 isomers (e.g., 2) have six aromatic
electrons.1
Minyaev and Gribanova proposed heteroatomic extensions of 1 with
eight-membered ring perimeters (Figure 1, 3, X = NH and O).2
We confirm these to be minima, despite having eight, rather than
six
electrons (see Figure 14S for the
MO's of 3a).
In contrast to this limited number of phC's, Wang and Schleyer predicted numerous molecules with planar pentacoordinate carbons (ppC) in 2001. They showed how ppC chemistry might offer unlimited possibilities for generalization and experimental realization.3 Appropriate ppC structural units can be grafted onto virtually any arene or unsaturated ring having three adjacent CH groups. Likewise, generally applicable construction principles allow the incorporation of phC structural units, based on simple elaborations of 1 and 3. The preparation of numerous planar tetracoordinate carbon compounds following theoretical predictions4 offers hope for the realization of phC's.
Akin to principles devised for ppC's,3 building block strategies facilitate the design of numerous molecules with phC's. The initial idea was based on the finding that C2v CH2B62- (4), the 1,2-dihydro derivative of 1, retains the phC despite its ruptured BB ring bond3 (albeit with somewhat elongated CB bond lengths). As in the design of ppC molecules,3 the opened edge of 4 can be bridged and the ring closed by appropriate atoms and groups. The selection of phC examples in Figure 2 retains the planar hexacoordinate carbon CB6 unit. Like 3, more than one bridge is possible. Consequently, phC's can be grafted onto myriad systems.
Figure 2 Examples of phC minima optimized at B3LYP/6-311+G**.
Bond distances are shown in Å, the lowest frequency ( min) in cm-1,
HOMO-LUMO energy separation (Gap) in eV.
|
The planar hypercoordinate bonding principles for phCs, like those for ppC's,3 are general and are easily extended to other combinations of atoms, with elements other than carbon in the centers.2 Thus, Minkin et al. computed examples of planar hexacoordinate boron species in 2001,5 and the boron analogue of 4, C2v H2B7- (S-4a, see Figure 5S), was reported recently by Boldyrev and co-workers.6 As in earlier studies of planar hypercoordination,1,4 we focus here on carbon as the central element since violations of the conventional tetrahedral tetracoordinate bonding of carbon seem more unusual. While all our new phC molecules are local minima,7 their isomers (e.g., with boron in the center and carbon on the outside) can be lower in energy. However, such species are no less interesting inherently and illustrate the generality of the bonding principles.
The two hydrogens in 4 provide substitution sites. Ring closure,
for example, by replacing both the H's, as well as the two negative
charges, of 4 with a CH=CH bridge, results in the neutral six
electron C3H2B6 (5). The appreciable vertical ionization potentials
(VIP) of 5 and other neutral species imply stability (Table 1S).
The six Wiberg bond indices (WBI) to the central carbon in 5,
ranging from 0.46 to 0.80, document the hexacoordination to the
B6 ring. The total WBI (3.86) of the central carbon does not violate
octet rule expectations. Exchanging the central carbon with the three
unique borons resulted in isomers (with hexacoordinate borons in
the center and carbons in the perimeter) ranging between 18.5 and
30.8 kcal/mol lower in energy (see Figure 4S). However, a Born-Oppenheimer molecular dynamics simulation8 (BOMD, Figure 3S
for the trajectory), using the deMon 2004 program,9 demonstrates
the viability of 5 as a local minimum and its resistance toward
isomerization. Searches only located high-energy transition states
for isomerization.
Benzannulation of 5 gives 6, the 10
electron analogue of
naphthalene. Similarly, the replacement of vicinal hydrogens in
benzene and essentially all arenes by a CB6 group (i.e., 4 without
its H's) can yield new phC candidates. The tropylium ion derivative,
C8H5B6+ (7), exemplifies other bicyclic systems. The tricyclic
minimum, C2v C8H2B12 (8), has two CB6 units, each with a phC,
grafted meta onto benzene.
Elaborations of 4 with single-atom bridges result in phC minima with seven-membered ring perimeters. For example, a methylene group can replace the two H's in 4 and bridge the CB62- unit to give Cs C2H2B62- (9). While the methylene carbon in 9 deviates slightly from the CB6 plane, the planarity of the CB6 moiety is nearly exact. Replacing the two H's in 4 with CH- results in C2v C2HB6- (10); planarity is retained. Heteroatomic bridging groups such as NH can result in favorable neutral phC minima, such as CHNB6 (11).
The phC units in 4-11 have 4n + 2
electrons. Their
aromaticity is demonstrated by refined nucleus-independent chemical shift (NICS) indices, such as the perpendicular tensor components, 1 Å above the central carbons (NICS(1)
zz).13 The NICS(1)zz
grid of 5 and the large negative value (-45.0) above the ring center
indicate the presence of a strong induced diatropic ring current,
with a shielding zone inside and a deshielding zone outside (see
Figure 7S). Dissection of NICS(1)zz shows that the shielding tensor
contributions from the
MO and from the in-plane radial
MO
sets are both diatropic (NICS(1)
zz = -18.5; NICS(1)radialzz = -14.3.
See Supporting Information Figure 10S and Table 1S for NICS(1)
and dissected NICS(1) results). Schleyer et al. coined the term
"doubly aromatic"10 for such situations. The NICS(1)zz results for
5 indicate that both
and radial MO's help to achieve the planar
geometry with a phC. Planar hypercoordination may benefit even
more from
MO stabilization than from the
MO system. S-4a,6
with only four
electrons, also illustrates this point.
Two-fold bridging of 1 on opposite edges results in an eight-membered ring perimeter exemplified by Minyaev's phC molecules,
3 (X = NH and O), which have eight
electrons (see Figure 15S).
The isoelectronic 12 (i.e., D2h C3H2B62- where X = CH- in 3)
also has eight
electrons, but two of these (in HOMO-1) are
nonbonding due to remoteness of the CH units; the remaining six
electrons are delocalized in the central CB6 moiety (see Figure
11S; neutral D2h C3H2B6 is a triplet).
Replacing two H's in 4 with an exocyclic C=CH2 results in C2v
C3H2B6 (13), a phC minimum. Although 13 has a total of six
electrons, two of these are localized as a C=C
bond (HOMO-4;
see Figure 10S). Four
electrons remain for the CB6 moiety, and
a +66.2 paratropic NICS(1)zz value for 13 results. The C2v C3H2B62-
dianion (14), with six of the eight
electrons on the CB6 unit,
results in a -53.4 diatropic NICS(1)zz value, thus 14 may be more
viable than 13. Double C=CH2 bridging gives the eight
electron
phC minimum, D2h C5H4B6 (S-15; see Figure 5S for details).
However, akin to 13, S-15 has four
electrons localized on
exocyclic C=C bonds (HOMO-3 and HOMO-5; see Figure 14S);
only four
electrons remain on the CB6 unit. As expected, the 10
electron dianion D2h C5H4B62- (S-16; see Figures 5S and 15S)
has six
electrons on the CB6 unit and a diatropic NICS(1)zz
(-55.8). Molecules like 3 (X = NH and O), 12, and 14 demonstrate
that the total
electron count is less important than the number of
electrons associated with the phC moiety. Evidently, delocalized
bonding of the
system is needed to achieve planar hypercoordination.
CB6 units with planar hexacoordinate carbons (or borons) can
be grafted onto olefins, arenes, or saturated carbon systems. Most
of these new phC candidates are doubly aromatic with delocalized
as well as
systems, but a total of 4n + 2
electrons is not
required. All these phC derivatives provide additional examples of
deltahedral bonding involving carbon and offer many opportunities
for experimental realization.
This work was supported by the National Science Foundation Grant CHE-0209857 in Georgia, the 111 Project (B07012) in China, and Swiss National Science Foundation (Grant: PA002-115322 to C.C.). We thank Abdolreza Sadjadi for helpful discussions, and Frank Weinhold for an earlier version of the NBO 5.0g program.
Optimized geometries of phC and selected isomers, NICSzz grids of 5, CMO-NICSzz of selected molecules, MO figures, MD trajectory of 5, a summary of NICS(1)zz, and dissected NICS(1)zz results and Gaussian archive files. This material is available free of charge via the Internet at http://pubs.acs.org.
* In papers with more than one author, the asterisk indicates the name of the author to whom inquiries about the paper should be addressed.
University of Georgia.
Beijing Institute of Technology.
1. Exner, K.; Schleyer, P. v. R. Science 2000, 290, 1937-1940.
2. Minyaev, R. M.; Gribanova, T. N. Izv. Akad. Nauk. Ser. Khim. 2000, 5,
786-796;
3. (a) Wang, Z.; Schleyer, P. v. R. Science 2001, 292, 2465-2469.
(b) Li,
S. D.; Miao, C. Q.; Guo, J. C.; Ren, G. M. J. Am. Chem. Soc. 2004, 126,
16227-16231.
4. (a) Minkin, V. I.; Minyaev, R. M.; Hoffmann, R. Russ. Chem. Rev. 2002,
71, 869-892.
(b) Sorger, K.; Schleyer, P. v. R. THEOCHEM 1995, 338,
317-346.
(c) Röttger, D.; Erker, G. Angew. Chem., Int. Ed. Engl. 1997,
36, 812-827.
(d) Radom, L.; Rasmussen, D. R. Pure Appl. Chem. 1998,
70, 1977-1984.
(e) Siebert, W.; Gunale, A. Chem. Soc. Rev. 1999, 28,
367-371.
(f) Choukroun, R.; Cassoux, P. Acc. Chem. Res. 1999, 32, 494-502.
5. (a) Gribanova, T. N.; Minyaev, R. M.; Minkin, V. I. Mendeleev Commun.
2001, 169-170.
(b) Gribanova, T. N.; Minyaev, R. M.; Minkin, V. I.
Russ. J. Inorg. Chem. 2001, 46, 1207-1210
6. Alexandrova, A. N.; Koyle, E.; Boldyrev, A. I. J. Mol. Model. 2006, 12,
569-576.
7. All the phC's, optimized with Gaussian 03 (see ref 11) at the B3LYP/6-311+G** DFT level, were minima, and their wavefunctions were stable.
Magnetic shieldings-a good measure of aromaticity or antiaromaticity-were evaluated by nucleus-independent chemical shift methods (NICS,
see ref 12). In particular, out-of-plane (zz) tensor component contribution
from individual canonical molecular orbitals to NICS (CMO-NICS; see
ref 13) computed using the NBO 5.0g program (ref 14) at the GIAO/PW91PW91/6-311G* level was used to analyze the behavior of the radial
and
MO's.
8. BOMD was carried out on 5 for 10 ps after 2 ps of equilibration with the interval of 0.5 fs at 300 K at the PBE/DZVP/Gen-A2* level.
9. Köster, A. M.; Lores, R.; Geudtner, G.; Goursot, A.; Heine, T.; Patchkovskii, S.; Reveles, J. U.; Vela, A.; Salahub, D. R. deMon; NRC: Ottawa, Canada, 2004.
10. Chandrasekhar, J.; Jemmis, E. D.; Schleyer, P. v. R. Tetrahedron Lett.
1979, 39, 3707-3710.
11. Frisch, M. J.; et al.; Gaussian 03, revision C.02; Gaussian, Inc.: Wallingford, CT, 2004. See Supporting Information for the full reference.
12. Schleyer, P. v. R.; Maerker, C.; Dransfeld, A.; Jiao, H.; Hommes, N. J.
R. v. E. J. Am. Chem. Soc. 1996, 118, 6317-6318.
13. (a)
ernu
k, I.; Fowler, P. W.; Steiner, E. Mol. Phys. 2000, 98, 945-953.
(b) Corminboeuf, C.; Heine, T.; Seifert, G.; Schleyer, P. v. R.; Weber,
J. Phys. Chem. Chem. Phys. 2004, 6, 273-276.
(c) Heine, T.; Schleyer,
P. v. R.; Corminboeuf, C.; Seifert, G.; Reviakine, R.; Weber, J. J. Phys.
Chem. A 2003, 107, 6470-6475.
(d) Fallah-Bagher-Shaidaei, H.; Wannere,
C. S.; Corminboeuf, C.; Puchta, R.; Schleyer, P. v. R. Org. Lett. 2006, 8,
863-866.
14. Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J. E.; Bohmann, J. A.; Morales, C. M.; Weinhold, F.; NBO 5.0; Theoretical Chemistry Institute: University of Wisconsin, Madison, WI, 2001.