$CHOOSE Problems and Avoiding Banana Bonds

$CHOOSE Problems and Avoiding Banana Bonds

Postby wtransue » Tue Oct 10, 2017 5:18 pm


I would like to compare the polarization of the C...N pi interaction between diazomethane H2C=N=N and formaldehyde hydrazone H2C=N-NH2. I'm running into two issues here, one per molecule.

Diazomethane. If NBO is left to find its own Lewis structure, it comes up with H2C:–N≡N: with lone pairs on carbon and the terminal nitrogen. Since I'd like to compare the C...N pi interaction, I'd like to specify H2C=N=N:: with two lone pairs on the terminal nitrogen. When I use the $CHOOSE block to do so, I get an error "Linear dependence encountered; $CHOOSE structure is likely bogus at atom 1" (the carbon is atom 1). How can I surmount this issue?

Formaldehyde Hydrazone. Likely due to hyperconjugative effects, NBO finds two pseudo-banana bonds between the C and N rather than a sigma and a pi bond, which makes the comparison I'd like to do impossible. Can I coerce NBO6 into printing a sigma and a pi interaction rather than these two bent bonds? Using $CHOOSE doesn't seem to change anything since NBO is already finding two bonds; they're just bent instead of what I want.

Posts: 2
Joined: Sun Feb 14, 2016 8:00 pm

Re: $CHOOSE Problems and Avoiding Banana Bonds

Postby ericg » Wed Oct 11, 2017 7:53 pm

I encountered the same problem with diazomethane. A similar failure with CHOOSE (directed) searches was recently reported by another user, and I'm working on a fix for the problem. I hope to have a fix within the next week and will post new NBO6 distributions at that time.

Mixing of sigma/pi character in the CN double bond of formaldehyde hydrazone results from slight pyramidalization of the amino group. There are no options in NBO6 to prevent such mixing, particularly in C1 symmetry geometries. However, one of the CN bonds clearly has considerable pi character, and the other sigma character. You might use these bonds as is.

Alternatively, you might consider constraining the amino group to be planar. The resulting Cs geometry would force the separation of sigma and pi. I would bet that the polarization of the CN pi in this case doesn't differ too much from the polarization of the (predominantly) CN pi of the equilibrium C1 geometry.

Posts: 267
Joined: Sat Dec 29, 2012 9:31 am

Return to General NBO Discussion

Who is online

Users browsing this forum: No registered users and 1 guest