GC method with H3PO4 and KH2PO4 in the sample solution

[Brian1706]

Hello guys!

I am trying to develop a GC method on a GC-MS or -FID system with a wax coloumn.

My solvent will be an alcohol like 1-BuOH. Due to the sample preparation there will be H3PO4 and maybe KH2PO4 soluted in these solvents.

Could there be any concerns for the coloumn or the detector (MS or FID) when H3PO4 and/or KH2PO4 is present? If so, what amount of concentrations could still be acceptable?

Best regards!

[Don_Hilton]

You probably want to avoid gettign strong acids or bases onto a wax column - as this will shorten the column life. You may be able to catch much of the phosphate on glass wool in the inlet - but as you collect the phosphate in the inlet, you will be looking at an increasignly active inlet liner.

How much is too much? This depends entirely on the actual rate of degredation of the column and what you are willing to pay as the cost of analysis per sample.

Is this preparation your only option? I assume that you are extracting the analyte from a phosphate buffer with a butanol/water or pentanol/water extraction? If so it would be a somewhat polar analyte? What about SPE or even SPME? -- So you can avoid getting the salts into the inlet at all.

Effect on the detector - I don't expect the phosphate to get that far unless you have conditions where you will get some phosphate ester formation with the alcohols. In that case - we analyze phosphate esters by GC and GC-MS.

[Brian1706]

Thank you for your quick answer.

I am performing a hydrolysis and I use the H3PO4 to "neutralize" (pH 2-3) the KOH in the solution.

I read that Na ions are a problem in FID detectors as their formed oxides are polluting the detector. Will there be the same problem with K ions? And what about MS detectors?

The inlet has 260 °C and I am also running the oven temperature to 260 °C.

[rb6banjo]

The salts won't get to the detector so that's not a concern. They likely won't even get out of the inlet. However, they will build up in your inlet thus changing it's characteristics. As Don mentioned, you will likely start getting some unwanted activity in your inlet which will change your chromatography.

What are your analytes in the extract? Are they amenable to SPME or static headspace analysis?

[Brian1706]

Unfortunately, we do not have the possibility to use SPME or headspace right now.

So if the salts are not a problem, there still would be a H3PO4 concentration of about 0,1 mol/L in the solution.

Is it possible to estimate the rate of degredation of the coloumn caused by this?

[rb6banjo]

None that I know about. If the acid makes it to your wax column, I fear that you might notice degradation fairly quickly.

[Don_Hilton]

If you are injecting a largely aqueous mixture with 0.1 m phosphate, I do not see this as a good way to go. Even if you are taking an butanol or pentanol extract of this aqueous mixture, the alcohols will pick up a substantial quantity of the aqueous phase - including the salts.

I would seriously look at putting the mixture down a SPE cartridge and eluting with a strong solvent so the salts are left behind.

If you want to (and it does not disclose information that should not be disclosed) describe the analytes and matrix, perhaps we can give you some ideas that will help to avoid getting the salts into the GC inlet.

[Brian1706]

I am performing a hydrolysis of esters in a system that only contains an alcohol like 1-BuOH and KOH and no water.
The H3PO4 (85 %) is used to stop the reaction and get the pH to 2-3. After that I analyze the alcohols by GC-MS or GC-FID.

I could use an ion exchanger to get rid of the K ions, but I would guess that KH2PO4 will not be soluted by the solution anyway?!
But I will have H3PO4 in the solution, so that could be the bigger problem?!
In the end of the sample preparation there are 0.05 mL H3PO4 (85 %) in 20 mL sample solution.

[Peter Apps]

Isn't this a transesterification rather than an hydrolysis ?. Your products are butyl esters which would partition nicely into a non-polar solvent (unless there are polar groups on the acid end) that would leave the salts and phosphoric acid behind in the butanol. Some butanol would also dissolve in the non-polar phase, but you are injecting large quantities of butanol anyway so it should not be a problem.

Peter

[Brian1706]

With KOH in 1-BuOH you mostly get hydrolysis products and only small amounts of transesterification products.
Also, the question on whether hydrolysis or transesterification occurs should not matter for the analysis of the alcohols?!

[Don_Hilton]

If I understand, you start with a solution of KOH in butanol, run the reaction and add H3PO4 and have a solution with H3PO4 and salts - all in solution in the butanol. Even at 0.1 M, I would not put this into a GC system if I could avoid it. If you must do this, use a GC inlet liner with a large plug of glass wool - there is at least one type that has a plug at least 1 cm long. And be ready to change it frequently. Run a check standard with analytes of interest (but no phosphates) every few samples until you get a feel for how long you can go between changes of the inlet liner without getting poor results from an active inlet.

I'd rather get the salts out of the solution before making the injection. And partitioning into water is a good way to do that -- assuming your analytes will partition well into the organic phase.

Getting a clean separation that clearly separates water from butanol is tough. But you may be able to take a mL of your reaction mixture and partition it between 20 - 50 mL DCM and some water. Alcohols should be recovered largely in the organic phase and - assuming you are reacting suffient ester in this mixture, you should have strong enough peaks to get a good trace on the GC without introducting salts into the GC inlet.

Ion exchage can be used to get rid of both anions and cations - but without further information, there is the consideration of loss of analytes to the polymer that makes up the resin. If you want to go in the direction of a column separation, I would look at SPE, where you would wash away the ions with water and then strip organics from the column/cartridge with a solvent that would go nicely into a GC.

[Brian1706]

After a few experiments on the solubility of the salt (KH2PO4), I would assume that it is not soluble in 1-BuOH.

So, I will not deal with phosphates but H3PO4 in the solution. This will probably not stay in the inlet, but get onto the wax column. That is why I am more worried about the H3PO4 and the coloumn than the salts and the inlet...

[Peter Apps]

After a few experiments on the solubility of the salt (KH2PO4), I would assume that it is not soluble in 1-BuOH. If you have done the experiment you do not need to assume, however you do not need much solubility to have enough in solution to mess up your inlet

So, I will not deal with phosphates but H3PO4 in the solution. This will probably not stay in the inlet, but get onto the wax column. That is why I am more worried about the H3PO4 and the coloumn than the salts and the inlet...

How is phosphoric acid going to get onto the column ? Maybe dissolved in aerosol droplets, but a plug of glass wool in the inlet should fix that. It will still mess up the inlet though. .

I think that it is fair to say that the consensus so far is that your really really should not inject salt and acid solutions into a GC if you can avoid it. Is there any reason why you cannot use the clean-up methods that have been suggested ?

Peter

[Brian1706]

One of my analytes is ethylene glycol which should be highly soluble in water. If I am running SPE with water to wash out the ions, I would expect ethylene glycol to get washed out, too?! But I have to admit that I do not have many experience with SPE clean-ups...

I was just "assuming" the non-solubility because my experiment had a detection limit of 5 mg/L...

[Peter Apps]

One of my analytes is ethylene glycol which should be highly soluble in water. Indeed, and if you had told us this at the beginning the answers could have been a lot better focussed - the real challenge is to find a clean up selects between inorganic phosphates and ethylene glycol. Is there anything else that you are not telling us ?If I am running SPE with water to wash out the ions, I would expect ethylene glycol to get washed out, too?! But I have to admit that I do not have many experience with SPE clean-ups...

Why do you need to add excess phosphoric acid ? If you add just enough to neutralise the KOH the insoluble phosphate will drop out. How about adding calcium hydroxide to precipitate the excess phosphate ?

Peter