IPA/Acetone Analysis on 6850

[cawarhur]

Greetings community! I have a few application questions regarding the analysis of isopropanol and acetone by GC.

Our setup:

GC: Agilent 6850
Column: HP-5
Inlet Liner: Single taper with deactivated glass wool
Sample: Acetone, IPA, hydrocarbons

1) Should I worry about IPA "sticking" to the glass wool in the liner? If so, should I choose a new liner or derivatize? Because the sample is to be washed through silica gel prior to GC analysis, I was considering derivatization as a possible step for the chemist running the sample through the silica as a way to prevent the IPA from sticking to that as well.

2) I can't separate acetone and IPA for whatever reason on this column. Without getting a new column, derivatization with BSA or BSTFA is my solution. Neither derivatizing agent will affect an HP-5 column, correct?

3) If the silver bullet to my analysis is derivatization, is there a way to calibrate a GC using a derivatized sample, or is the error going to be too large to give meaningful results (i.e. is there anyway of quantifying the extent of the derivatization process)?

-My idea was to first calibrate the GC for IPA, then derivatize and check the cal curve for amount of IPA calculated and if it reads none, to say that when using BSA as solvent, the IPA is completely derivatized. When I make that conclusion, I can derivatize a few different concentrations of IPA and then use that calibration with my actual samples. This hinges on IPA making it through the liner wool obviously.

Thanks for the help. I always appreciate the comments I get from you guys as you've all helped me many times in the past. ^_^

[Consumer Products Guy]

1) Should I worry about IPA "sticking" to the glass wool in the liner? If so, should I choose a new liner or derivatize?

No.

2) I can't separate acetone and IPA for whatever reason on this column.

Both acetone and IPA are essentially unretained on HP-5, expected.

Without getting a new column, derivatization with BSA or BSTFA is my solution.

No, derivatized IPA will not be separated from the derivatizing agent, etc.

Neither derivatizing agent will affect an HP-5 column, correct?

That is true.

3) If the silver bullet to my analysis is derivatization, is there a way to calibrate a GC using a derivatized sample, or is the error going to be too large to give meaningful results (i.e. is there anyway of quantifying the extent of the derivatization process)?

-My idea was to first calibrate the GC for IPA, then derivatize and check the cal curve for amount of IPA calculated and if it reads none, to say that when using BSA as solvent, the IPA is completely derivatized. When I make that conclusion, I can derivatize a few different concentrations of IPA and then use that calibration with my actual samples. This hinges on IPA making it through the liner wool obviously.

Thanks for the help. I always appreciate the comments I get from you guys as you've all helped me many times in the past. ^_^

I just don't see this working for you. You need a column with polarity to seraparate the acetone and IPA. Here we use Rtx-624 or PEG columns for small alcohols.

[cawarhur]

I was able to obtain GC traces of acetone and IPA on this column, albeit within a few minutes of injection. I also derivatized IPA and saw what I loosely assumed was the peak for the silylated IPA, separate from the derivatizing reagent which makes me think that your statement that I won't see the derivatized product is incorrect here.

In the mean time, I'll take a higher mw alcohol, derivatize with BSA, and note whether or not the peak I thought was the silylated IPA is still present. If it is some side product, I should see that same peak in this GC trace, too. If not, I think it is safe to assume that I can, in fact, see the derivatized product.

Thank you for answering my other questions, but I'd like some other perspectives, too.

[HbJ]

So I guess IPA and Acetone are the traces you're looking for in the hydrocarbon solution, right? What boiling point/range does it have? What concentrations do you expect?

And what length/ID/film thickness etc. does your column have? This is crucial.

For now, my answers to your questions are as follows:

1.) IPA does not "stick" on the glass wool. It's much too low boiling to do that.

2.+3. Your silver bullet is optimization of your separation first.
a.) What carrier gas are you using? If you have an EPC capable unit, optimize for the perfect linear velocity.
b.) If possible, use EPC and nitrogen. This _is_ a silver bullet.
c.) Use a low oven temperature, like 40°C.
d.) Reduce your injection size to avoid overlapping peaks.

A chromatogram would be really helpful also.

[cawarhur]

How could I have forgotten such important bits of information?

So I guess IPA and Acetone are the traces you're looking for in the hydrocarbon solution, right?

Yes.

What boiling point/range does it have?

BP range is roughly 56°C-160°C

And what length/ID/film thickness etc. does your column have?

30m x 250um x 0.25um

a.) What carrier gas are you using? If you have an EPC capable unit, optimize for the perfect linear velocity.

Helium with no EPC

c.) Use a low oven temperature, like 40°C.

Oven program is starts at or very close to 40°C

d.) Reduce your injection size to avoid overlapping peaks.

The chromatography looks fine--sharp gaussian peaks--and we're only injecting 1uL and then splitting it off.

I'll get a chromatogram up here asap! Thanks again!

[Johnny Rod]

As said above, a 624 or other PEG type column is what you need to separate IPA and acetone. Even if you get slightly different RTs when injecting the single components, you probably won't get useful separaiotn of a mixture. It's a lot easier to get a new column than to have to derivatise! Plus you are adding a source of variation in there by adding more steps.

[HbJ]

I just looked up the RIs for acetone and IPA on an HP5 and I don't see any possibility for a good separation, at least with helium and such a small film thickness.

As a note: You're using an unpolar thin film column meant for high-boiling components to separate low-boiling, polar solvents.

The only options I can think of right now is to use nitrogen and optimize linear velocity for an oven temperature of 40°C if you want to just separate Acetone and IPA.

As this would be just tinkering mostly, I'd really recommend a thick-film 624 or WAX column, as suggested.

A last note: Please save helium and use either hydrogen or nitrogen. It's a finite resource and people running ion trap MSs (like me...) are in desperate need of helium.

[Consumer Products Guy]

The smallest molecules we can typically resolve from derivatizing agents are ethylene glycol-2TMS and propylene glycol-2TMS. Note that those are 2TMS derivatives.

[cawarhur]

Thanks for all the help everyone. I'm looking into a new column. The problem now is I need one that will separate acetone and IPA...and also in the sample is C8's isomers (cyclooctane, cyclooctene, and COD).

Looks like I'll need two columns, right?