[Babcock_Hall]

Enthalpy,

Thank you.  The data set that you provided is very interesting and difficult for me to explain using simple models.

[Enthalpy]

That puts my mind at rest, because I spent days just trying to find general qualitative rules with disappointing results. For sure, broadly known rules like "more branched" don't work.

Symmetry is one factor, for instance cubane and adamantane stay solid at heat. Understandable at molecular representation: if a liquid molecule arrives to the solid surface and can stick optimally with 6 orientations like cubane, this favours the solid. This explains well that one or two branches favour the liquid, especially if they're not at the molecule chain's center, but many branches may favour the solid.

Rotation freedom also favours the liquid as independent parts of the liquid molecule can have orientations unfavourable to stick to the solid surface. Thermodynamics would tell this in an abstract way. And this must explain why tetramethylsilane stays liquid when neopentane freezes, while the boiling points differ little, suggesting similar intermolecular forces.

But for the rest... I fear any prediction really needs a brute software that finds the optimum stacking by trying all orientations and offsets, computes intermolecular energy, and compares with the heat contents of the liquid. That's what the research groups try to do if I grasped it properly. Less pleasant than qualitative rules or additive ones (which don't work that little bit), but very useful if the software works some day.

Mischievous Nature has foreseen solids where cubyl groups keep their rotation freedom.

[YouSheng]

Although no conclusion, thanks for all reply and data provided by Enthalpy