Chromatography Forum

Hello!

I am trying to collect some calibration data for the GC, and I have been having a hard time getting rid of the residues from my previous analyses. For example, I inject 1 ul of 10% acetic acid in water, and do a blank run right after. I do see NO peak in my blank run(s). Then I inject distilled water. (I use FID so I shouldn't see any peak.) And there it is! My acetic acid peak... Its area is 1/10 of its area in my original analysis. If I continue injecting distilled water its area decreases gradually. Other compounds that I injected also showed the same result.

Here is the information on the system I am using:

GC: HP 5890II
Column: AT-Q
Column flow: 4 cc/min
Aux flow: 26 cc/min
H2 flow: 30 cc/min
Air flow: 400 cc/min
Total flow: 100 cc/min.
Septum purge: 3 cc/min.
Oven temp: 40 to 180 at 15 C/min
FID: 200 C
Inj: 200 C
GC equipped with Electronic Pressure Control system

I am using an HP autosampler to inject my samples and I tried having it wash the syringe 30 times with my solvent (water), 15 times with my sample (again water), and pump 15 times with my sample. I am pretty sure this is not an issue with the syringe. I am thinking my sample is getting trapped somewhere in my injection port and with my next sample being swept into the column. But I don't know for sure. Has anyone ever come across with such a situation? By the way, there is an on-line valve installed on my carrier gas line.

Thanks,
Hakan

Acetic acid is being washed from its resting place when you inject water. One has to discovered where it is resting. It can be salted out on a cationic site like bare metal or deposits in your injection liner or column, or it can be found upstream of your injection point if your gas flow is not fast enough to overcome the 'backflashing' of the water injection as it is vaporized in the injection port.

Water increases its volume 1300 times when vaporized at 200°C.

The speed of your injection and the amount of water you inject has a direct bearing on the possible overfilling of your injection liner volume.

If the acetic acid is 'salting' out downstream on your column there is little you can do unless you replace the column or remove a length of the front of the column.

best wishes,

Rod

I had a similar problem with the separation of acetic/butyric/pentanoic acids on a packed column. Turned out to be that we were using liners packed in-house using glass wool that was not completely deactivated.
Either remove your inlet liner packing or replace it with one pre-packed with deactivated glass wool.

After reading the replies here I found out that the liner in the GC had non-deactivated glass wool. So, I ordered deactivated one and replaced the liners. I can still see peaks in my water injections but thay are way smaller than the previous case. And my analyses are now reproducible.

Thanks a lot for your comments!

Hakan

Hakan,

It will also help if your autosampler is set for a slower injection rather than a fast injection. You were probably depositing HoAc upstream to a cooler place in the pneumatics and it took time to flow back downsteam toward your injector where the acid could then be washed out by another injection of water.

The active wool was 'catching' the acid and holding it for a later injection.

With a slower injection the water is less likely to deposit the acid upstream. This should improve your carryover issue. As you have found the wool replacement did help a great deal

best of luck with your work.

Rod

Rod-Supelco used to sell quartz wool for applications like this. Do you still sell it and have you any evidence that it works? When I did this with packed columns I used to remove the wool entirely with good effect. I am not sure what injection technique hakanolcay is using and what effect the wool he is using is having-besides being an adsorption site for his anaytes. So I would not suggest removing it at the moment-especially if the analysis is satisfactory

Supelco sells this wool as a custom product, that is, it is not listed in the catalog, but it can be purchased if someone calls or emails the company and requests a quotation. It is also available in custom packed columns.

Hakan's GC has an oven temperature of only 40°C. He is probably using the default injection speed of the HP autosampler which is very fast and which intensifies the effect of the explosive expansion of the water in the 200°C injector.

The porous polymer particles glued to the capillary tube may retain the HoAc to a degree, especially if contaminated from previous injections of samples(septum particles and other involatiles). The HoAc can flow back down through the injector and the column after the injection deposits it upstream in the pneumatics if the injection liner is unable to completely hold the expansion volume of the injected sample.

This at least can happen. Since he was using undeactivated wool, the wool is a likely site for retention of this backflashed and downloaded acetic acid and other analytes from the injected sample.

Quote: "Other compounds that I injected also showed the same result."

Replacing it seemed to help. The acid could have been left from the injection or redeposited from pneumatics bleed.

Reducing the speed of the injection allows the water to take more time to expand and assists in minimizing the amount that may be backflashed into the pneumatics, if that is happening. This is a common cause of variability in GC analyses.

Hope I am clear in presenting possibilities of the cause of the problem and how it can be minimized.

best wishes,

Rod

Rod- well it seems that it is still possible to buy quartz wool, but you didn't offer us any data suggesting that it is indeed better to use it in this application and that it really gives better results than "deactivated" glass wool. It appears from what you are saying that few people are using quartz wool, otherwise it would not be a custom order.

I am no clearer on the injection technique that is being used here. I would recommend cold capillary on column injection for this analysis with a small volume injection. At least this technique eliminates the possibility of adsorption in any injection ports. But if split injection is being used, I was wondering if using an inverted cup liner or something like that might be preferable to something that contains wool. Of course, a number of different problems might be introduced here such as incomplete sample vaporization, so I hesitated to recommend this, especially if the method is now working acceptably......

Victor,

Perhaps I made implications I did not intend to make.

You said "you didn't offer us any data suggesting that it is indeed better to use it in this application and that it really gives better results than "deactivated" glass wool "

I did not intend to make that claim and I believed I achieved that goal.


GaryR stated that he found that to be the case with his similar analysis. I do not wish to deny his experience.

Nor did I suggest that the wool should be removed.

I did intend to imply that it may not be possible to eliminate all carryover of acetic acid due to the kind of column Hakan was using.

And there is a difference between deactivated wool and quartz wool. Deactivated wool is not made of quartz. And it may not be as inert as quartz wool.

Quartz wool is brittle, unlike deactivated glass wool, and is often even more inert than deactivated glass wool. I hope this is clear.

My only recommendation was concerning the speed of the plunger of the needle, the injection speed. In the autosample used with the HP 5890II the default injection speed is extremely quick, in the blink of an eye. A slower speed (one or more seconds in length) will decrease the amount of flashback in the injector.

best wishes,

Rod

Joining back to the conversation... I hate to say this but after about 100 injections I started seeing large peaks again with water injections. But I am not complaining because the analyses are still reproducible -don't know how..

Maybe I should try quartz wool, or maybe I should first try removing the glass wool from the liner. I checked if I could adjust the speed of the injection but I couldn't find a way to do that (the autosampler is a little old, HP 7673B, it may not have this option). But you are right, it is like the blink of an eye.

On the other hand, like I said, every time I am injecting 1ul (this was the smallest that I could get with the autosampler) but the thing is I am doing splitless injection (also not cool on-column injection).

Thanks again for your replies.

Hakan

Maybe the deactivated glass wool became activated again after being acid-washed (i.e. with my injections)...

With the high temperature acidic water (steam), it certainly is a possibility, almost a certainty.

best wishes,

Rod

ps Quartz wool might be worth a try

Do you have access to another autosampler/GC that has the capability of performing slower injections?

Unfortunately, I don't have access to any other autosampler. But I will certainly try the quartz wool.. I ordered already.

Thanks,
Hakan

hakanolcay,

The injection speed, if my memory serves me correctly, can be set from the dip switches inside the front cover of the injector although I may be confusing this with on-column injections. Also, regarding 1 uL injections, if you replace the 10 uL syringe with a 5 uL syringe, the first stop on the autosampler will only inject 0.5 uL instead of 1 uL. 5 uL syringes commonly available from reputable GC supply houses.

Good luck.

Unfortunately, the dip switches do not control the injection speed. They are only good for the volume, injections per vial, and controlling the washes. The on column injection setting is made from a board inside the injector. However this setting also increases the syringe travel by 19 mm so if you do not have the on-column needle guide installed, the syringe will bottom out and the run will crash. Or so I have heard.