May simply be a function of thermal history. Check if the cooling scans are identical.

Thanks for the response Peter. I typically only use 1st heating cycle to compare thermal histories.

On the cooling scans, both display only a very small exothermic peak around 100°C, mainly due to the fast cooling rate (10 ° C/min). The onset of crystallization peak on cooling start 3-4 degrees earlier (higher temp) on Batch B than Batch A. I also noted that the cold crystallization temperature for both batches shift up to 108-109°C on 2nd heating cycle, which would indicate the difference I saw in the first process was most likely due to thermal history. However, I'm still unsure why batch B with a higher molecular weight (same PDI and % crystallization) would crystallize at a lower temperature on first heating cycle? I'm assuming there is another factor contributing to cold crystallization temperature that I haven't considered. I had thought about molecular orientation caused by the extrusion process, but both batches had identical elongation at max load, with batch B demonstrating slightly higher elastic modulus and tensile strength.

Dear Brian,

I am wondering about whether the relationship of "higher residence time in extruder yielded lower Mw" for the sample A, is based on an experimental data or a theoretical assumption. That might not be the case. Because, you have also pointed out that PDI did not changed. You may simply compare the zero-shear rate viscosity values of specimens to check effect of melt processing on Mw. If it is true (B>A), you should consider chain dynamics in solid-state which are obviously different from those in melt-state.

Thanks for your response Ali.

As determiend by GPC, the Mw reduced from 419,715 (dried resin) to 367,575 for batch A and 413,439 for batch B. I was actaully quite happy with the little amount of molecualr weigth loss I got during both processes.

I had considered chain dynamics, but not sure how exactly they are reducing my cold crystalisation peak? very little in the literature on it.

Thanks

Brian

Hello Brian,

I would start with basic checks of sample mass and repeatability before going too far into this. If the sample masses are similar and this experiment is repeatable over two or three samples of A and B then you have something.

I would then consider the possibility of the degraded polymer that exists in batch A hindering the crystallization process. This will not be the same as having excess monomer or smaller Mw chains, for example, as chain scission during processing produces reactive radicals. This will change the properties of the extruded polymer without necessarily changing the PDI and %crystallinity since it's only a few percent of the chains.

Good luck.

Hello Brian,

Roddel Remy's answer makes sense to me.

The cold crystallization is occuring well above the glass transition (as expected), so the chains in the amorphous phase are quite mobile, at least in a localized way. Batch B with the higher molecular weight crystallizes sooner, both on heating and on cooling, than batch A. In other words, the higher MW amorphous phase is less stable against crystallization in both cases. If molecular weights were much lower I would say that's as expected anyway (longer chains --> higher probability of them aligning themselves into nuclei). As things stand, with modest MW difference between A and B and very low ratio of chain ends to chain bodies, it is likely that the presence of degradation byproducts plays an important role in inhibiting nucleation.

But I would definitely make sure that the results are repeatable, independent of sampling etc. The results are not miles apart. If you could test a batch that underwent even more thermal degradation than batch A you could be more certain that the trend is real.

Step 1. How to do reliable DSC measurements:

Please do 3 heating/cooling cycles, with the highest temperature above the (highest) melting temperature. Please have a slow cooling rate unless you are preparing a "glassy sample".

1. First cycle may be useless as it is expected to include non relaxed stresses/thermal history.

2. Theory says that the second cycle should be the right one but you do need a proof.

3. Third cycle provides the proof unless you did not started the degradation of the sample. The expectation is that the second and the third cycles are identical.

Thanks for all your responses, it provides me with some additional areas to focus my research on. I plan to do some chemical analysis and determine if this is higher degree of degradants in batch A versus batch B.

Just to clarify, I have ran numerous samples with a mass between the 8 to12 mg range and all showed the same trend. Also, this phenomenon is only present on the first heating cycle, as both batch A and B have almost an identical Tc on the second heating cycle. Thus indicating the difference may be related to thermal history.