Well, they wouldn't have to be made with precision provided there's an easy way to separate them. Which is still difficult but could be doable.

> Moreover the difference between 1.05 and 1.06 nm drastically affected freezing point.

I think the author might have misread something. The Van der Waals radius of a carbon atom is only 0.17 nm, so the radius of a single tube can't vary in steps of 0.01 nm. Maybe this just referred to the average tube radius?

Hmm. A tube is composed of many carbon atoms in a ring. Add one more atom, you increase the circumference by 0.34nm but the radius only by 0.34/2pi=0.054nm. So still looks infeasible.

Perhaps at this time.