I have checked a injection molded PP sample by DMA which was first cooled to -100°C then heated to 150°C and cooled again to -100°C at 1°C/min heating and cooling rate and at 1 Hz. In the heating step I can see a Tg of roughly 2-7°C (in tanDelta) while in the cooling step right after I find a much lower Tg between -8°C - -4°C. My question now would be: is this change due to a change in properties (crystallinity, etc) of the PP at high temperature or is it normal that you have a shift in Tg depending if you analyse heating or cooling? Is it because of disordering or thermal history of PP sample during heating? Or it is only because of slippage occurred during DMA analysis?
Can you post the thermogram? Once you heat something over its TG you erased its thermal history and characteristics. You said that on the cooling step you notice a TG. I dont think you can quantitate TG on the cooling step, thats crystaliization thats happening.
I am attaching here the thermogram. Thanks Alex to help me to find solution.
Alex is right to say that heating something above Tg erases part of its thermal history. Tg is a function of heating/cooling rate. If a sample is cooled quickly, the polymer chains have less time to rearrange themselves and the glass transition occurs at a higher temperature; the result is a glass with a high free volume. If a sample is cooled slowly, the polymer chains have more time to move and rearrange themselves even at lower temperatures; the glass transition temperature is pushed lower and the result is a glass with less free volume. The equivalent effects are seen on heating.
If your Tg measured on heating is higher than that measured on cooling, the implication is that the cooling rate you're using in your DMA experiments is slower than the rate at which the component was cooled during manufacture. You could possibly confirm this by doing another DMA run on the same sample; the second time, you should see the lower Tg on both heating and cooling steps (assuming the heating and cooling rates are the same!)
For further information, see any good text book on polymer science or (for example) the paper by Moynihan et al., J. Phys. Chem. 78 26 (1974).
Thanks for your answer, i suggested almost this explanation but one of our scientist was not satisfied with this answer, rather he gave another suggestion which i mentioned in my question and i had no paper to explain him. Now i can explained him very well by taking help of this paper.
As I hinted in my earlier answer, there are ways to test whether my explanation is the correct one. I suggest performing a second DMA run on the same sample, using the same heating/cooling rate. You should then record the same Tg on both heating and cooling steps.
You could also try to measure the density of the sample before and after DMA. If my explanation is correct, the sample with the lower Tg should have less free volume and be more dense. Whether this effect is significant enough to measure by any practical method, I can't say.
Hopefully these hints will help you to establish the truth and convince your colleague!
Thanks, i asked to my colleague to run second time with the same sample and we can measure the density of these samples by using Davenport gradient column (ASTM D 50568) Method.
Dr. Howie,
What about the possibility of increasing the crosslinking when heating the polymer above the Tg? Might you expect the second run Tg to be slightly higher than the first run heating ramp? I'm not as familiar with interpreting cooling curves as we only look at heat curves in my lab and we typically see higher Tg on the second run. However, our method does include a rapid cooling which supports your point that a fast cooling rate will lock the strand in a more chaotic array with higher Tg. I noticed there is no minimum in the rubbery modulus as would be seen by the phenomenon of rubber elasticity. Do you have a comment about the effect of heating on crosslinking and the calculation of cross link density from DMA modulus curves?
Best Regards
Hello Mrs. Levine,
It depends on which kind of cross-linking and density of cross-linking. Sometimes stereo-chemistry also plays the role in increasing the Tg . I also want to know if somebody knows that how to calculate cross-link density from DMA modulus curves?
Regards
Felicia,
If there is any additional crosslinking at high temperatures, it would indeed increase Tg. I'm afraid I'm not expert enough to guess which effect will be dominant. Probably you could separate the two by modifying the cooling rate. If your increase in Tg is due to crosslinking, it will not be affected by cooling rate.
Best,
Phil
Hi Bhawana , have you already solved this problem? I also encounted the same problem. Could you share your method to solve the problem?
- Badges
- Science topic
- Similar topics
- Geoscience
- Atmospheric Sciences
- Climate Science
- Climate Change