## Lesson 3 -part 6 -vid 6; Lesson book p.71 - Q6

mcatacct1
Posts: 65
Joined: Sat Oct 26, 2019 4:35 pm

### Lesson 3 -part 6 -vid 6; Lesson book p.71 - Q6

which of the following is closest to the O-Si-O bond angle in silica?

The correct answer is 109 based on the text "each silicon is bonded to four oxygen atoms". I chose 180 degrees based on the molecular formula provided "SiO2" and its description as a "homolog" to CO2.

I see justification for 109 and 180. But I decided to go with 180 because it is consistent with the molecular formula provided in the passage. What information am I missing, or what info am I not synthesizing here to arrive at the right answer?
NS_Tutor_Mathias
Posts: 616
Joined: Sat Mar 30, 2019 8:39 pm

### Re: Lesson 3 -part 6 -vid 6; Lesson book p.71 - Q6

Right, it is not described as a geometric homlog, so what kind of homology is meant is left a little bit vague. You are better off considering the network solid with each silicon bonded to 4 oxygens.

This is for the most part pretty cut and dry, and I think you've extracted all the relevant information here. Don't get tripped up by non-committal statements like that cursory mention of homlogy between CO2 and SiO2. This is probably meant to say that they are both dioxides with group IVa elements. There is after all no way to have 4 identical groups bonded to a single central atom and maintain a 180 degree angle between each.
mcatacct1
Posts: 65
Joined: Sat Oct 26, 2019 4:35 pm

### Re: Lesson 3 -part 6 -vid 6; Lesson book p.71 - Q6

I guess my problem stems from the fact that I thought molecular geometry was always related to the molecular formula (SiO2). Should I not make that connection? If I should make that connection then what are the exceptions to this besides network covalent solids?
NS_Tutor_Mathias
Posts: 616
Joined: Sat Mar 30, 2019 8:39 pm

### Re: Lesson 3 -part 6 -vid 6; Lesson book p.71 - Q6

The formula definitely doesn't guarantee molecular geometry. Lone pairs are the most obvious case of departure from such a prediction, but in general you need to be fairly certain of the nature of those bonds in the structural formula and of the electronic configuration of the central atom.

That said, prediction electron geometry from total number of bonds and lone pairs works fairly well (with the caveat still that you need to know about all the lone pairs. This can be tricky in some cases where the octet rule is violated).