If you suspect that your soil samples contain appreciable quantities of inorganic C (i.e. CaCO3), it is necessary to remove inorganic C for obtaining accurate measurements of 13C isotopic signature of the organic C, since CaCO3 has a different 13C isotopic signature.  You could do a simple preliminary experiment to analyze several replicated samples of your soil WITH and WITHOUT Carbonate removal treatment applied to exactly check how your results will be affected by the presence of CaCO3 in your soil.

A useful reference explaining steps for Carbonate removal in soil samples can be found in the attached publication.  Hope this would help you.

Drying, grinding and acid treatment

Hi, Yuan.

Fumigation with HCl in a closed dessicator is the choice. As to the time needed, amount of sample, etc., again I suggest you take a look to the bibliography.

The (excellent) paper of Ramnarine et al. (2011), uploaded by Dr. Jayasundara (see above), proposes to do this in glass vessels. After destroying carbonates, weigh the appropriate (decarbonated) amount of sample in tin capsules and perform elemental analysis + isotope ratio in your equipment. The problem is that the destruction of carbonates involves a mass loss: thus you must correct for this, in case you want to obtain by this method not just the d13C, but also the % OC of your sample. If you are interested only in the d13C, then the correction for weight loss is not needed.

However the most widely used method is to fumigate in the capsule herself. You weigh the amount of soil sample in the capsule, and perform the fumigation before closing it. However, in such a case you must use silver capsules, because the HCl vapors destroy tin capsules. Silver capsules are more expensive; if you have a big lot of samples to analyze, this may affect your budget...

Take a look to this paper for more details, including time of fumigation:

Harris D., et al. (2001). Acid fumigation of soils to remove carbonates prior to total organic carbon or carbon-13 isotopic analysis. Soil Science Society of America Journal 65: 1853-1856.

Some of my colleagues use this method, but without fumigation: they add the HCl (100 microliters, or more if necessary) directly on the silver capsule. This simplifies the procedure; however you must be careful as to avoid losses due to strong efervescence, when samples are very rich in carbonates. Nevertheless, this approach would have an obvious advantage: you got a true control about how much HCl do you apply to your samples, and if you know previously their carbonate content (Bertrand's calcimeter, for instance?) then you may be sure that you add HCl enough to remove all CaCO3. Note that when using the conventional fumigation you may be quite unsure about how much HCl does really reach each individual sample, and also whether or not you have really destroyed all carbonates. Note that in mediterranean countries, such as my own one, soil horizons having 40% of more carbonates (by weight) are not uncommon.

Of course the main problem in all these approaches that use HCl is the need (compulsory) of removing all remaining HCl from the samples: if still present, it may cause corrosion problems in the machines (CN analyzer, mass spectrometer...).

Good work!

If carbonates are present (soil emits CO2 bubbles on addition of 1 M HCl) I suggest using acid, but a weak acid such as sulfurous.  There is literature on this in related to C measurements on soils using combustion analysers.