Polymers extension unclear

[merkys]

This issue has been discussed before, however somewhat inconclusively.

I would intuitively expect C&1&1 meaning a self-bond, i.e., a bond with another repeating unit. Thus, C atom constitutes a monomer with one outgoing and one incoming bond, hence the valence of 2. So I understand C&1&1 as depicting an endless saturated alkane chain. So the polystyrene example is clear to me.

However, I cannot understand the diamond and graphite depictions. Based on the previous example, in c&1&1&1 there are two outgoing and two incoming bonds at each c atom, resulting in valence of 4. Or am I missing something?

Moreover, I cannot locate Daylight (Weininger) proposal for this extension, which is referenced in the OpenSMILES specification. It would be interesting to understand this notation, as it seems quite an elegant way to denote polymeric structures.

[johnmay]

Looks like a mistake Polystyrene should be

c1ccccc1C&1C&1
c1ccccc1C(*)C* |$;;;;;;;_AP1;;_AP1$,Sg:n:0,1,2,3,4,5,6,8::ht|

but

c1ccccc1C&1&1

is not wrong per-say

c1ccccc1C(*)* |$;;;;;;;_AP1;_AP1$,Sg:n:0,1,2,3,4,5,6::ht|

Which would be polymer like this:

*C(c1ccccc1)C(c1ccccc1)C(c1ccccc1)C(c1ccccc1)*

[johnmay]

Those are CXSMILES BTW