TGA is one of them more advanced techniques

Thermal Gravitimetric Analysis



Thermal Gravitimetric Analysis (TGA) and Differential Scanning Calorimetry (DSC) are two ways of examining how stuff changes in responses to heat.

TGA measures how matter decomposes and changes in response to heat.  DSC is testing the amount of heat neccessary to change the heat in the sample.

Neither of these things is easy.  Simply heating something on a scale and balance would not work for a number of reasons.  First of all, the mass would be unlikely to be correctly placed and so would lead to unusual readings, the same way that the bathroom scale reads differently when you move around on it. So a good TGA array isolates the sample on a fine, and supported, scale.  This lets the changes in mass be carefully tracked.

Second, there are times when you want to separate the effect of the heat from the effect of a reaction.  If you heat something in air, it’ll burn, but that is the result of a reaction with the surrounding oxygen in the air, rather than a result of the changes brought by heat.

Changes that are a result of heat on their own are called pyrolysis.

So a good TGA should allow you to keep the materials in different gas environments, to separate the effects of pyrolysis from those of reaction.

Finally, a good TGA should have the option of linking up to a gas analyzer – a Mass Spec or whatever.


Essential for loading and separating the substances
TGA crucibles

The material under investigation needs a container.  There are several reasons for this.  First of all, you need to be sure that there isn’t contamination and carry over from previous experiments (I’ll write a note on how to clean TGA crucibles later).  Second, it makes loading the TGA more convenient (imagine if you had to dose them each time).  Third, it lets you measure the mass outside of the TGA before insertion.  Fourth, different kinds of crucible have different properties.

Pictured are ceramic crucibles, but for the Mettler-Toledo machine (pictured above) the following types are used:

  • Alumina.  The standard, basic ones.   Re-usable.
  • Aluminium.  Give a better signal – possibly due to temperature transmission – but they will melt at 660C
  • Platinum.  More expensive, but give a good signal.  The crucible will fail at 1600C.  Watch out for compatibility – carbon doesn’t like platinum and some things will form allows, messing up the readings
  • Sapphire.  Not actually the blue stone, they look closer to the white ceramic above, but more translucent.  Sapphires and Rubies are aluminium oxide, taking their distinctive colour from chemical impurities.  Sapphire is quite the standard, and works well for metals.


Sample and Sizes guidelines

  • Organics: 5-10 mg
  • Inorganic: 10-50 mg
  • Things that explode: 0.5 – 1 mg


Preparation of samples

Best samples are flat disks and powders.  If something is chunky, grind it up into a fine powder, but be careful – too strong grinding can cause chemical changes, not to mention include impurities.

Threads should be chopped finely and wrapped in aluminium foil.

And in general, the sample should be representative of the whole – of whatever you are preparing.  So it is, as always, advisable to do several readings.  Personally, I’d say that three are the bare minimum.

Loading the crucible


I have little to add to that video.  However, make sure that the material is flat on the floor and evenly spread.  Sealing depends on what you are looking for – if the substance is volatile and you don’t want the gas to escape, seal it.  Otherwise leave it open.  Ditto if you are testing the reaction with air, oxygen, carbon dioxide or whatever.

Gas Atmospheres

As mentioned, gases can be several different types.  My preference would be to test pyrolysis first – with everything – before going on to the more complex, entangled measurements.

Blank curves

A good blank will be with an empty crucible, preferably the one you are going to use.  Or rather, the ones.  If you have three crucibles, each serving as its own blank, you should get a nice data.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s