Non-Integer Form

Crystalline ultralong alkanes are normally found in one of the integer folded forms: extended, folded exactly in two, three etc. However, real-time small-angle X-ray scattering (SAXS) experiments using synchrotron radiation have shown that the initial melt-grown crystals usually have a fold length intermediate between those of the integer forms. This transient non-integer form (NIF) soon transforms into an integer form - see figure =>. NIF owes its name to the fact that lamellar thickness is not an integer fraction of the chain length,

Below is is a typical series of SAXS spectra recorded during cooling of the long parraffin C246H494 from melt. Pure NIF is observed between 120 and 111oC.

Fourier reconstruction of electron density profiles normal to the lamellae are shown in the figure on the left for two different temperatures. The maximum corresponds to the crystalline core of the lamella and the minimum to the amorphous intercrystalline layer. The thickness of the core is equal to that in once-folded integer lamellae. It remains constant while the thickness of the amorphous interlayer decreases.

The model of a NIF lamella and its density profile are shown below. While some chains are folded, others are not and traverse the crystal layer only once. This leaves long chain sections (cilia) uncrystallized, forming the amorphous layer. With time the cilia find their way into the crystal, resulting eventually in highly crystalline material with once-folded chains.

The reason that NIF forms at all is that it crystallizes fast. New chains deposit randomly on the side of a growing crystalline lamella (Figure (a) below). If regular once-folded crystals (c) were to form directly, all chain ends would have to be positioned at the crystal edge before deposition. In contrast, the growth of NIF (b) is not hampered by this constraint and is therefore rapid. However, the price paid is the low crystallinity of NIF. Subsequent conversion of (b) into (c) proceeds by solid state diffusion. A similar process takes place in "secondary crystallization" of polymers,

Supporting the above mechanism is the behaviour of long alkanes with a short branch (Me, Bu) half way along the chain. Since the branch does not fit into the crystal, already the initial deposition of the first half of the molecule must be "correct", i.e. with the chain end and the branch at opposite edges - see (d). This means that the second half of the chain is already in a correct position to crystallize completely, without the need for chain translation - Figure (e). As expected, NIF is barely observed in such branched alkanes and the integer once-folded form (f) appears almost instantly.

References

Long Alkanes

Group Homepage