Possible problems and solutions

There are many different calorimetric techniques and methods of use (open crucible, closed crucible, tight crucible, work under different types of gas, under pressure etc.) and consequently a number of different problems that may be encountered by the user. Below are some possible solutions according to the particular problems.


Possible problem



The sample contains moisture

If the sample is heated in an open or non-tight crucible, an endothermic peak corresponding to the water vaporisation will be seen. This can mask the thermal effect of the transition or reaction to be detected. To prevent water vaporisation, work in a sealed and tight crucible. Be careful with the vapour pressure in the container. Remember that the water vapour pressure is 100 bar at 300C.


The reference contains moisture.

Mostly alumina powder is used in the reference crucible to balance the heat capacity of the sample.
It is recommended to firstly heat in a flame the alumina powder to remove moisture, in order to avoid any thermal effect of vaporisation on the reference side of the calorimetric detector.


Deformation of the crucible bottom after sealing

According to the sample loading of the crucible there may be a deformation of the bottom of the crucible. When using a plate-type DSC, the contact with the container will be poor and this will produce a drift of the baseline. The solution is to prepare a new crucible with less sample.

Metallic sample

The interaction of a molten metallic sample with the DSC metallic crucible

Embed the metallic sample in alumina powder so as to avoid contact with the metallic crucible when melting.

Solid organic sample

The sublimation of the solid organic sample

Some organic samples (e.g. benzoic acid) sublime during melting. To measure the temperature and heat of melting of such organic samples, it is necessary to work in a sealed crucible.


The baseline drifts

If the baseline drift is coming from the sample determination, there are two ways to correct for it:
  record a blank curve (with empty crucibles) keeping the same experimental conditions; then subtract the resulting curve from the sample curve
  have a second sample run with the same experimental conditions and subtract this curve from the sample curve (this is valid if there is no reversible effect during the second run)


The baseline is noisy

Heat the DSC without crucibles, in air, up to 600C in order to burn organic residues that can disturb the detector. Then run a blank test (with empty crucibles) to check the baseline. If the baseline is still noisy, a check-up of the instrument is needed.