I have an analytical apparatus that measures the total carbonate in solid materials by acidification of the material under a stream of $\ce{CO2}$-free air, followed by quantitative coulometric titration of the liberated $\ce{CO2}$. The coulometer has a detection limit of about a microgram of $\ce{CO2}$, and so long as the material can be finely ground and homogenized we will typically use a sample size of 2 mg. The sample is acidified with 5 mL of 2 M $\ce{HClO4}$ and heated to $\mathrm{80^oC}$ while sparging with the $\ce{CO2}$-free air stream.

For materials like $\ce{Na2CO3}$ or $\ce{CaCO3}$, the carbonates are thoroughly liberated as $\ce{CO2}$ and analyzed with full recovery within 5 minutes. For materials containing significant amounts of dolomitic limestone ($\ce{CaMg(CO3)2}$) or siderite ($\ce{Fe(CO3)3}$), the analysis times can approach one hour or more. This is problematic not just because of the reduced sample throughput, but also because the detection limit of the coulometer increases with the length of analysis due to the time-integrated noise at the detector.

Although the issue described above has been a problem with this analysis for decades, it had always been given low priority until now. Over the years we have tried different non-oxidizing acids (nothing volatile like halo-acids, as they would be titrated as $\ce{CO2}$), propietary commercial catalysts, and organic wetting agents. The configuration described above has given the best results (lowest time to complete recovery of carbonate), even though in worst-case scenarios (i.e. low total carbonate, predominantly present as dolomite or siderite) the results are essentially useless and all we can do is report high detection limits.

I suspect that we probably can’t come up with a better acid than perchloric: it’s a very strong acid, it is not oxidizing under the conditions of this analysis, it’s non-volatile and perchlorate is stable and it doesn’t tend to form precipitates that might occlude portions of the sample, etc.

My thoughts are that there should be some reagent that would displace then complex the $\ce{Fe}$ and $\ce{Mg}$ in siderite and dolomite, respectively. We do not want to oxidize any organic carbon that might be present, however (at least not all the way to $\ce{CO2}$).

So, how can I alter this analysis so that the carbonates of dolomite and siderite are more rapidly liberated as $\ce{CO2}$, without oxidizing any organic carbon that might be present, and without introducing acidic gases to the coulometric detector?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.