Hi Pierluigi, 

If the Tan d rises and falls gently that's quite likely to be real, but if a flat then a sudden jump, that does sounds a little odd.  I think I'd add the strain onto a Y axis to see that it was being controlled successfully the whole time and also look at temperature vs time to check no temperature control switching had occurred at around 90°C.

Hello Pierluigi, based solely on the information that you provided, I tend to believe that this is not an intrinsic property of the material (since abrupt changes are not as common in a rheological experiment), but your measurement might be approaching certain physical limits of the configuration of the experiment (geometry size, frequency used) including reaching limits of the instrument (minimum torque, motor inertia).  These might ultimately lead to your experimental conditions to be less optimized when your material gets very low in viscosity due to increases in Temperature, where it was optimized when the test was at lower temperatures.  With that said, this could result in the rheometer/geometry configuration possibly having inertia affects when measuring bitumen-like material at 90°C where the material is very low in viscosity (" G" quite linear") and you get an abrupt change in G'.  IF your phase angle is close to 90° before the abrupt change, then it could mean that your measurements are instrument inertia dominated.  One of the best ways to determine if this is the case, is to plot "raw phase angle" (might be a different variable name based on the rheometer manufacturer), however if this value nearing 180° or above 170°, then the inertia correction being applied by your rheometer is significant (even if you have calibrated the inertia of the geometry and instrument), and your sample response is therefore being masked by the significant inertia correction, which might result in "abrupt" changes in the calculated G' and G" that relies on the "corrected phase angle".  I would suggest that you contact your rheometer manufacturer's helpdesk and provide them the data file for them to investigate your experimental conditions under the higher temperatures, to suggest improving the experimental setup.  This might result in a different geometry, a different frequency, a smaller gap, etc.  If the experimental conditions are determined to be fine in that temperature range, then further investigation of what could cause an abrupt change and back to the same value will need to be conducted.  However, as you are doing, rule out the instrument configuration and experimental conditions first.

Regards,

Fred 

Dear Pierluigi,

That is a bit strange, did you do one test?, several tests?,apart of checking your test conditions (eg., gap, geometry) if you had that plot while doing onte test it could be something related to your machine (a noise for example), if you had the same trend while doing several tests, then you have to check if you had a "phase transition" or "structure disruption" at that temperature, also, you have to find out if your sample shows changes in its structure such as Tgs (glass transitions), because changes in moduli can be seen during such transitions, and some materials may show several Tgs (different temperatures).

Regards

Javier

Dear all, thank you for your answers. I have forgot to say that I am performing experiment on plate-plate geometry. This sort of peak is quite constant for all the experiments that I have performed so far, no matter the amplitude. According with DSC results that show a little endothermal phenomenon around the same temperature, I am about to think that is just due to evaporation of volatiles, perhaps water absorbed from the environment. I have recorded a really slight mass loss. By the way the peak is realli abrupt, and is located between 90-100°C. The weird thing is that is more intense than how much should be expected for such amount

What I can get from this is that you are running a temperature ramp? what range?, are you using a sovent trap?, perhaps mineral oil would be of some help to avoid some evaporation of your sample, what seems bizarre to me is that some volatile compounds are responsible of the storage modulus decay, I thought if this happened  it was the other way round. If you are having too much sample evaporation, you have to consider that you might be changing your rheological system (big change in tandelta) , that means that as the test goes on, you no longer have the original sample you started with.

Hi Pierluigi,  I think you're right that the loss of volatiles is giving the dramatic change...  it is possible that the sample is being partly expelled before the surface tension pulls it back under the plate.  If as you say it's moisture loss at 90-100C, you won't really be able to prevent that without either pre-drying the sample or using a high pressure cell.  The vapor pressure of water above boiling is significant.

http://slideplayer.com/slide/1424209/4/images/33/Vapor+Pressure+of+Water.jpg

Great for Philip, that should be another factor that you have to take into account Pierluigi..

Regards

Thank you Philip. Indeed my compound is a complex mixture highly viscous and hygroscophic. I have tried to remove the water but is not possible in vacuum at moderate temperature, and i can't push the temperature further without have some reaction in the sample. I had now the TGA-MS confirm. I should find some way to have a better-looking graph to be pubblish

Great Pierluigi, it is good to know that you are on your way to sort it out....