Hello Nazli,

Based on my experiences, to plot the master creep curve, you can look into either the time-temperature superposition principle or the time-stress superposition principle.  Both of these principles are based on the same free volume theory, which can be used to construct the master creep curve from short-term creep tests.  From what I know, most of the short time creep tests are carried out on mechanical equipment which is very easy to do.  With varying temperature or creep stress values, a set of short-term creep-strain data can be used to determine the temperature or stress shift factor.  Depending on the polymer that you are working with, the relationship between the shift factor and variable can be linear or non-linear.  Thus, the generated master creep curve can also behave linear or non-linear viscoelasticity.  Hope this can help.

Good luck.

Bing

As Bing said, it should be possible, but there are a couple of potential problems about TTS:

1. thermal degradation of your material might occur faster than you anticipate, which will falsify any TTS attempt

2. depending on the material you are using you might change the structure of your material when heating it. This includes changes of crystalline structure (melting), ageing (shifts Tg and does not happen in a way that it can be shifted with TTS), and evaporation of solvents and/or low molecular substances.

In principle, I would prefer a DMA over a tensile tester, if you can allow for testing long enough to get appropriate creep data, i.e. if the DMA is not overbooked. You have a better chance to do the tests in N2-atmosphere than in a tensile tester, as a modern DMA has this feature built-in.

I would agree with using the time temperature superpositon principle as a first hypothesis. The use of DMA depends on the available frequency range. Usually it is not too broad. If you can prepare a reasonable J' - J" curve (i.e. the compliance master curve not the complex moduli!) you can apply invrse Fourier transformation to get the creep compliance curve vs. time. Or, alternatively you may try to use TMA (thermomechanical analysis) with small loads at different temepratures and you can prepare the creep curve using the time-temperature superposition principle directly, without frequency-time transformation.

I appreciate your kind responses however the question is the comparison of DMA method to mechanical testing with the aim of plotting master curves of specific polymeric materials. Are there any disadvantages that using DMA rather than doing mechanical testing equipment may bring ? Such as for example the frequency range as Professor Banhegyi states ? which one is more reliable ? which one is more efficient ? which one is the fastest method, or does it take the same amount of time to plot a master curve with both of these methods ? My aim is to plot master curves of polymers such as PC, PE, PMMA etc. thank you.

Kind regards

Nazli

Hi Nazli,

Two points:

1. You need to consider which mechanical loading you want to see reproduced in the DMA, together with the frequency/temperature sweep you want to examine. But which viscoelastic engineering constant you want to investigate? Bending, tensile/compressive, or shear?

2. Master curves are numerical interpolations, you need to have a sufficient number of points to obtain a good R^2 fitting.

Take care,

Fabrizio

Hi Nazli,

I misread the post - I thought you were considering classical master curves in the frequency/temperature domain. I think Professor Banhegy points are quite valid. Regarding whether it is best to use a DMA or mechanical testing, it may depend upon the type of DMA you are using. I personally find most DMAs 3P/4P bending fixtures less reliable than the analogous you can find in mechanical rigs. Bending is also a complex deformation field, uniaxial test may provide a better option, but it depends also on the type of application you are looking for your polymers. Tension/compression are generally OK, shear in my experience best done with pendulum/torsion, however there are some pure shear fixture that also work well.

Best,

Fabrizio 

Hello Fabrizio,

Thanks for your answers, very helpful indeed. In a nutshell I am aiming to plot master curves that explain me the long term behaviour of polymers under constant strain (stress relaxation) or constant stress. Considering these are general material characteristics, the samples will be tested under various tensile stresses and temperatures to later extrapolate these curves and line them up to plot the material characteristics in long term - for example 1M cycles. Hence, I will only study them under tensile loads.

Best regards

Nazli

Dear Nazli, If I understand well, you want to simulate repeated loading cycles, which complicates the situation further. Simple time-temprature superposition is not enough for fatigue prediction, see e.g. http://www.eng.uc.edu/~beaucag/Classes/Characterization/DMA%20Lab/IF1.1%20Fukushima.pdf

http://barbero.cadec-online.com/papers/2010/10Guedes(Ed.)ISBN978-1-84569-525-5Chapter2Barbero.pdf

http://journals.sagepub.com/doi/abs/10.1177/0021998308093913?journalCode=jcma

http://www.hanserpublications.com/SampleChapters/9781569904725_9781569904725_Understanding%20Polymer%20Processing_Osswald.pdf

https://sem.org/wp-content/uploads/2015/11/sem.org-2006-SEM-Ann-Conf-s82p03-Formulation-Time-temperature-Dependent-Fatigue-Strength-CFRP.pdf

What the significance of Master-curve (TTS) from

1. Three point Bending

2. Dual cantilever

3. Single cantilever

@Nagamadhu M: theoretically, it doesn't matter, with which geometry you measure the data. Practically, clamping will always lead to artefacts, which is why, I would prefer 3-p bending, if possible (if the sample is not too soft)