direct water inject

[mr3573]

in order to determine aromatics such as benzene and ethyl benzene in water matrix, i inject 1 ul of sample directly to wax type column. for calibration of this method i should make and inject mixture of few ppm of each component, but the repeatability of results not so good. how can imporove method?

[Don_Hilton]

The repeatability problem could be the result of any number of details.

How are you making up your standard mixture of a few ppm comonents in water?

What GC conditions are you using. (Inlet tempreature and initial column temperature can be quite important.)

What does the chromatogram look like?

And, while you did not ask: Water on a wax column can be a problem in that water may result in degredation of the column, particularly if there are ionic or highly polar compounds dissolved in the water. For analysis of benzene, ethyl benzene and a number of compunds of similar volatility, there are a number of analytical techniques around that avoid getting water onto the GC column. Purge and trap comes quickly to mind. SPME or other solid phase extration techniques work nicely. And extraction with a solvent such as dichloromethane does a nice job of giving an extract with nice chromatographic characteristics.

Many published techniques have excellent reproducability.

[Consumer Products Guy]

I inject 1 ul of sample directly to wax type column.

You don't detail, but if it's a capillary column I'd use 0.5 ul injection, especially if you have a 0.32mm column or smaller. 1 ul can overflow the injection liner.

[chromatographer1]

Your problem with reproducibility can be attributed to the fact that water coelutes with benzene. It also would be broad enough to affect the sample plug on the column concerning toluene and C8 aromatics.

You should use another column in front the wax column or replace it.

best wishes

Rod

[mr3573]

Thank you for all reply. The basic method was based on extraction by pentane and inject of organic phase. But it was so difficult on routine analysis (4 or 5 time in day) extraction procedure. So I found a literature on net about direct injection of water to wax type column without major degradation on column. New program have these specification. Inject temperature:150 / split ratio 1:20 / oven temperature:85 / FID temperature 200
Column is capillary 0.32 . for making standard I added 25 ul of each organics (benzene, ethyl benzene) to one litter of water which is equal to about 20 ppm of each. I heard that head space could improve precision and repeatability. I haven’t any experience to use headspace. Is that useful for routine analysis and can improve it?

[chromatographer1]

Headspace can measure accurately, plus or minus routinely, 1% error, on a range from % levels to ppb, depending upon the analyte, but almost any volatile solvent down to 1 ppm with a dedicated headspace analyzer, PE, Tekmar, Agilent, and others.

Using a simple 5% phenyl column joined ahead of your wax column should solve your problem.

OR

You could use a capillary or packed porous polymer column to resolve the water issue. A one foot packed column should be enough but 3ft might be better. Micropacked would be better still if attached to a inlet liner with sufficient volume to handle the volume expansion of the water.

SPME will bring down the detection limit another 3 orders of magnitude if you need to go there. I have done benzene in water with a Tekmar HS analyter, Model 7000, down to 100 ppt with a Varian 3500 GC-FID.

But 1 ppb is easily attainable.

Good luck,

Rod

[Spuzzin]

For aqueous injections I normally use on-column with a 5m 0.53mm ID precolumn. Tends to improve repeatablity.

Rich

[chromatographer1]

With cooler initial temperatures this tends to 'collect' the liquid water but allow the aromatics to continue down the column and make a sharp plug at the column when the hydrocarbons reached the liquid phase of the analytical column.

A good suggestion, similar to putting a silicone phase column in front of the wax column.

best wishes,

Rod

[crosemeyer]

Looking to set up a similar method for ethylene glycol.

Using a Restek Stabiliwax column, 30m, 0.53mm, 1um. HP 5890 with S/SL inlet.

Their literature reccomends a Uniliner for direct injection--we don't think the boss is going to spend $65 on a liner. Can I get away with split or splitless injection? Any suggestions on ratio? And what about gooseneck vs. straight liner?

Thanks for any help.

[chromatographer1]

Dear crosemeyer,

Amazing isn't it. How much does your time cost? The overhead for the lab, wasting one hour, two, or maybe three trying to get good results because you don't want to spend $65 for parts to do the job well and correctly. Does this cost $65, or more than $65?

And then either accepting poor answers or no answer because gosh, it is just too expensive to continue to waste time.

Do you have a cold on-column injector?

Can you do the analysis by HPLC? refractive index, isocratic, with a column that can handle 100% water mobile phase?

Well, let your boss make the decision. If he doesn't want to listen to a vendor whose BUSINESS it is to make customers happy, then maybe the analysis isn't worth doing.

Or just write down an answer and turn it in to him and say: "I got it done cheap. Cost me $0.10 for a pencil. Here are your results."

If he is happy, then go on to the next task at hand.

Good luck,

(sorry about the excessively sarcastic remarks but it is amazing how easy it is for some bosses to waste time and money which are 'free' on the books, abut which give YOU ulcers and the blame when the results don't make anyone happy)

Rod

[Spuzzin]

Crosemeyer,

I struggled to develop a glycol suite (aqueous injection) using split or splitless injection. In the end I switched to using the restek direct injection liner, mainly because it was cheaper (short term) than converting to on-column. Like you I had multiple 'discussions' with my manager regarding the cost of the liner. In the end it was the time it was taking to put together a reliable method and the argument of "let me get one to test" that won him round (grudgingly) in the end. It's not perfect but it's reliable once you get the hang of installing the column into liner. My advice would be to install the column into the liner while it's hot, I get a much better seal that way.

Rod,

I'm very sure your remarks made many of us think "thank god it's not just me". Made me smile as well

Rich

[chromatographer1]

Dear crosemeyer,

I am glad to see Rich confirmed my opinions with you.

After you get the liner installed, heat it to about 200°C or so.

Make a clean cut on your 0.53mm ID column. Wipe it off with a dripping chemwipe soaked in acetone. (this allows the polyimide coating to soak up acetone softening it, and making it sticky.)

Then in one motion, firming pressing up on the column, press the column into the outlet of the liner, not crushing it, but not allowing it to fall down from the inlet for several seconds.

Now you have a good seal as the acetone evaporates from the polyimide and it sticks to the liner. Don't allow anything to tug on the column when the column may be pulled off the inlet.

If the column comes loose, cool and remove the liner and clean the outlet so no particles might ruin your next attempt to seal a column to it.

I have had great success with these. Worth every penny.

best wishes,

Rod