[ScholasticSpastic]

I work in a commercial lab doing equal parts R&D and QC.  My background is a BS in Biology with a minor in Chemistry.  Due to oddities in my academic experience, I've wound up doing a lot more analytical chemistry than anything else.

The Director of R&D has asked me to look into the possibility of using industrial grade argon for ICP-OES rather than the UHP grade we are currently using.  My initial response was to make a very unflattering face and recommend against it.  However, in the spirit of fairness, I feel I should at least ask around to see if it's workable and what problems we might expect to face were we to try it.

Her background is mostly Biological, so it may be that she lacks the analytical experience required to see immediately that this is a very bad idea.  Please keep our respective backgrounds in mind as you respond to this post, should you choose to do so.  By all means use jargon on me, but if you think it might be difficult for me to translate from Chem jargon to Bio jargon, please consider using smaller words.   

Thank you so much for your time!

[Hunter2]

I think it has to be explained, that in ICP you have low determination values of elements. An industry argon is used for welding for example. It contains still some nitrogen and oxygen and carbon dioxide. Depending on the analysis you want to make it can interferre the results.

It the same to use Di-water or very clean double distilled millipore water for other analytical purpose. The grade of cleanness is important.

[Arkcon]

The real issue the lack of standardization.  Not only will low grade argon contain other elements, but the amounts will vary from time to time.  Once you spend time -- that's you, spending time, still getting paid, but not analyzing samples (you know, doing work, as he would call it) trying to alter instrument and sample handling parameters, the next tank could be different, and make all your work go to waste.

Not to make this personal, but I seem to face a similar problem.  The bosses would like to use tap water, instead of DI water for some applications.  Its a bad idea, as the pH may drift over time, with no explanation.  But this sort of money saving plan is the sort of thing like to call "random acts of management"^TM

"Can we use tap water"

Arkcon makes a strange face

"You have to think outside the box"

Arkcon makes a strange face and sputters"Bu-bu-but ..."

"You're not a team player, Arkcon"

"Have you ever even tried it?"

[Corribus]

I imagine this would be a bigger issue for ICP-MS than ICP-OES, given the nature of the types of interferences you are likely to encounter. Nevertheless, in a technique in which identifying potential contaminants can be exteremely difficult and costly, why would you want introduce something else to worry about? I mean, let's say you get an erroneous result. Now you've got to troubleshoot your gas on top of everything else. In an academic or government lab, this kind of potential headache just isn't worth the money you'd save.

Commercial labs of course have different priorities than academic and government labs, though, so maybe the risk justifies the financial reward in this case. If I were you, I'd talk to the instrument manufacturer and find out what they recommend. They will probably have a good idea as to what the risk is of using low grade argon. More than likely, it will also depend on what your potential analytes will be. (E.g, maybe it makes no difference at all if you're looking for lead, but can be a real problem if you're measuring cadmium.) Your instrument manufacturer's applications specialists should know this kind of thing.

[marquis]

I like Arkon's answer sooo much better.

You know murphy's law.  Here's the way it could work in
your case...

The ICP test start failing.  Because of this, the QC testing
gives failing results, meaning product can't be released.
Depending on your plant, it may require you to participate
in a bunch of programs like root cause, CAPA (corrective action,
preventive action), etc.

By the time you identify the interference, you may have had
to qualify an outside lab, develop new methods for the same
test, revalidate the method, etc.  All are very time consuming
and painful.  Been there, done that.

You will have interferences even with all the reagents of proper
quality.  No need to make things worse.

Good luck with this one.