Quick simple H-bonding question

[bomba923]

*Must a hydrogen be 'directly' bonded to an N,O, or F in the very structure of the molecule for it to engage in H-bonding with other such molecules?

*For example, can formaldehyde form H-bonds? --even though its hydrogens are not directly bonded with an oxygen in the structure? (i.e., not structurally attached) What about acetone or acetal?

*Or, must I concretely have an (O-H) or (N-H) bond within the molecule structure itself?

So, must a hydrogen be 'directly' bonded to an N,O, or F in the very structure of the molecule for it to engage in H-bonding with other such molecules?

 

[Artermis]

Taken from

http://www.chemguide.co.uk/atoms/bonding/hbond.html

:

"In methoxymethane, the lone pairs on the oxygen are still there, but the hydrogens aren't sufficiently + for hydrogen bonds to form. Except in some rather unusual cases, the hydrogen atom has to be attached directly to the very electronegative element for hydrogen bonding to occur."

 

[bomba923]

. .. Except in some rather unusual cases, the hydrogen atom has to be attached directly to the very electronegative element for hydrogen bonding to occur .. ..

Meaning formaldehyde, and possibly acetal, I presume ?
Not acetone?

 

[GCT]

Yeah, it's not a clear cut case. You're first assumption is correct in most cases, the hydrogen has to be bonded to the electronegative atom, this pertains to a local dipole moment due to the direct electronegative differences, which will facilitate intermolecular attractions so deemed to be hydrogen bonding.

You should also consider the scenario of hydrogen donor/acceptor capabilities when considering the nature of hydrogen bonding in a system.

And as artemis has mentioned there are cases where the partial charge differences between two hydrogen bonding candidates can be "stabilized" by hyperconjugation of electron donors and such.