Silicon Speculation

I would suggest signing up for the Plasma-Chem list server out of Syracuse. My first take is that open beaker digesting with HF is causing loss of Si as Silicon Fluoride gas. The closed microwave vessel traps it. Heating glass in an open beaker with HF is a great way to remove all the Si spectral interferences. The plasmachem list has many very experts in various realms of elemental analysis watching... When you say elevated levels that gradually go down, is this during the reading of a sample? ie rep 1 to rep 4 of the sample goes down? Thanks to J. Batchelor for providing this response.

Dear Sir, Thank you for your reply. We compare our values to ARMI, NBS or BCS standards which are evaluated not only using wet chemistry, but with spark-OES and XRF. Since the values of the verifiers are known before we run them, I started preparing the verifiers as I would a sample, by burning it to red hot in a propane torch. This is done to some samples to remove coatings or contaminates. When the samples or verifiers are not burned, and digested in teflon beakers with teflon covers on a hot plate, our values match very closely with what is expected. The HCl and nitric acids are added when placing the sample on the hot plate, and the HF is added after the sample appears to be in solution and has been removed from the hot plate. When the samples have been burned, the Si value can be double or triple the expected value for 2-3 days, and yes, over several readings, the value decreases to what is expected. These readings are always compared to a verifier or standard. The process of bringing it to the expected level can be excellerated by shaking the sample in the volumetric flask and exposing it to air helps, but I prefer not to do that since I am concerned about evaporation. I'm not concerned that I am losing a significant amount on the hot plate versus the microwave since the values never drop below what is expected, they are only elevated. Your thoughts?

This is definitely a question for some alloy gurus. [url]https://listserv.syr.edu/scripts/wa.exe?A0=PLASMACHEM-L[/url]sign up for an account. There are some rules...include an email signature block to show where you are from....its a list of scientists...vendors follow it so someone from PE might chime in. otherwise you will hear from others doing what you are doing. You can search the archives to see if anything has come up before, its worth it. James D. Batchelor,Ph.D. [url=http://www.waters.com/]Waters Corp.[/url] 1. Maple Street Milford,MA 01757 Thanks again to J Batchelor for providing this response.

We see elevated Si signals when we initially run samples digested in HF; the HF seems to leach Si out of the sample introduction system from previous samples that may have had Si but at a lower HF concentration or no HF (ours is an HF resistant sample intro system, so it is not coming from any glass components, but rather Si that has absorbed from previous samples). It seems it is necessary to precondition the sample intro system with a real sample to reach a stable Si signal, and how you prep the sample will have an effect on the active residual HF concentration (i.e. microwave vs open vessel). I suspect primarily the tubing as the culprit but we have not done studies to confirm this. Hope this is helpful. Thanks to S. Norris for providing this response. [url=http://www.vhglabs.com/]VHG Labs[/url]