[silverslothmaster]

I'm going through some papers about HSQ (hydrogen silsesquioxane) and keep seeing this fractional notation being used: HSiO_3/2 , SiO_4/2 , H₂SiO_2/2 , H₃SiO_1/2.

Example paper: https://www.physik.hu-berlin.de/de/x-ray/publications/pdf/caster_kowarik_jramanspectroscopy_2009_hsq_cars.pdf

I understand that the proper formula for HSQ is [HSiO_3/2]2n. Because we can't have fractional atoms, it's always multiplied by an even number and as such the fraction is never a consideration. But then why are the fractions maintained in the other compounds, and not reduced to SiO₂ and H₂SiO ?

Does keeping the fraction confer some structural information ? I've tried to find out what this notation means, but could not find a proper explanation.

I guess what I'm trying to find out is, what do the numerator and denominators mean, formally, in those formulas ? If it's that the oxigen atoms share outside bonds, why not write the compound as SiO₄⁻⁴ ? If the numerator is the number of atoms, what does the denominator represent ? And is it always /2 ? Can we have /x , where x is a different integer ? I've never seen this type of notation before. Does the denominator mean that the valence electrons are halved ? If you look at the linked paper's figure 1 where they draw out the cage, there's only one SiO₄ entity. The oxigen atoms are not shared with another SiO₄ entity, they connect to a H₃SiO_1/2, or a HSiO_3/2. I am really confused by this notation because I've not seen it in any textbook, and it's not formally defined.

Would it be correct then to say that silicon dioxide SiO₂ is made up of SiO_4/2 entities in which all oxygen atoms are shared with others ? Where 4 stands for four oxygens, and 2 means that each oxygen is shared between two SiO₄ entities ?

I've also asked this question on chemistry stackexchange but haven't gotten very far with my search for an answer. https://chemistry.stackexchange.com/questions/151392/what-does-fractional-notation-represent-in-molecular-formulas-such-as-sio4-2-or

[mjc123]

I'm not familiar with that notation, but looking at the paper it seems that H_aSiO_b/2, where a + b = 4, means a Si atom bound to a H atoms and b O atoms (each of which is shared between two Si atoms). Writing it that way emphasises that each Si is 4-coordinate. It's also not multiplied by an even number because it refers to a non-regular, partly crosslinked structure in which there are Si atoms with different environments, so these formulae refer to particular types of Si environment, and are not empirical formulae for the substance as a whole.

[silverslothmaster]

I see, thank you. Would it be fair then to say that the /2 is used to show that the O atom is a bridging ligand ? If so, can these be re-written as HSi(μ₂−O)₃, Si(μ₂−O)₄, H₂Si(μ₂−O)₂, and H₃Si(μ₂−O)? Or is that wrong because these are not empirical formulae for the substance as a whole ?