Chemistry Reference
In-Depth Information
1. Halogen bonding is largely an electrostatic phenomenon. Therefore, when
the halogen is bound to a more electronegative element, it becomes a bet-
ter electron acceptor. Conversely, base strength parallels (at least to the
first order of approximation) the effectiveness of a given species as a halo-
gen bond electron donor.
2. Halogen bond electron acceptor ability decreases in the order I > Br > Cl.
Fluorine (acting as an electron acceptor) forms only very weak halogen
bonds, if at all.
3. When a halogen is in close contact with another atom approximately 180
from the halogen-halogen sigma bond, it is acting as an electron acceptor.
When the contact is approximately 90 ,itisactingasanelectrondonor.
4. Formation of a halogen bond at one atom of a dihalogen does not preclude
the second atom from also acting as an electron acceptor. The stronger the
electron donor, however, the less likely that a D
···
X-X
···
Dcomplexwill
be formed.
5. Halogen bond formation results in partial occupation of a
σ antibonding
orbital, weakening the bond. Depending on the electron donor strength
and the environment, this can result in heterolytic fragmentation of the
bond, leading to ionic products.
While the general features of halogen bonding are now well known, it has
proven challenging to develop models with sufficient accuracy to predict
spectroscopic features and bond energies. This is particularly problematic
with iodine, where high quality basis sets are not readily available and are
computationally expensive. There have been numerous approaches taken to
address this issue during the past decade, many of which are discussed below.
Some surveys of computational methods involving charge-transfer com-
plexes of dihalogens have recently been reported. Alkorta and coauthors
looked at F 2 ,Cl 2 ,Br 2 , FBr, FCl, and ClBr interacting with several
σ
π
elec-
tron donors and using both the B3YLP DFT and MP2 methods [152]. Two
basis sets were used in this study: a 6-31G was used with the DFT method,
while a 6-311++G ∗∗ basis set (which includes diffuse and additional polar-
ization functions) was used with both the DFT and MP2 methods. Where
possible, the calculated halogen bond geometries were compared with experi-
mental values.
MP2 calculations gave better halogen bond distance estimations than did
DFT, with the latter method consistently underestimating the bond length
with both basis sets. The electron donor X-Y angle was found to be nearly lin-
ear in all cases, with a maximum deviation of 8 in the case of HF
and
Cl 2 .Bond
energy calculations followed the trends noted above, and the results corre-
late with hydrogen bond strengths with the same electron donor. However,
there were significant variations in the values predicted from each approach.
This is not surprising, given the different equilibrium geometries used; how-
ever, the correlation with bond distance was not consistent. The authors noted
···
Search WWH ::




Custom Search