Internal standard and surrogate response

[xie]

I've been following this posting closely because I'm having the same problem with a 5975A and purge and trap system (Tekmar Velocity). I did some liquid injections with an autosampler. The response was all over the place so I really couldn't draw any definitive conclusions as to whether it's the purge & trap or the MSD or both - couldn't even blame that on my own injection technique.

Have you made any progress?

xie

[AICMM]

krshaffe,

Simple thing to try. Bring your total flow down to 40 mL/min, bring your split flow down to 40:1 so you are using less than a mL/min. Set the instrument to constant pressure mode. Repeat the low point and high point of the curve.

Best regards.

[krshaffe]

Xie,

No I have made NO progress at all. I'm sorry that you are experiencing the same problem because it is really frustrating. I'm starting to think that it is something in the quantitation...in the software. Maybe something with the ions that it is using to quantitate. I just don't see how active sites could cause this problem. I don't know... I'm going to try what AICMM suggested. Let me know if you find out anything.

Thanks,

[Peter Apps]

Hi Katie

Can you give us a brief listing of what you have tried so far - as far as I can see it is limited to changing inlet components and fixing some leaks in the gas system. Have you done anything to the P&T, which is most likely where the problem lies ?

Peter

[krshaffe]

Hey Peter,

In the very beginning I suspected the P&T unit, but it's a brand new unit and I don't really know what to check on it. I'm used to the OI Analytical concentrators with the key pads on front where you can control it from there...and bypass the autosampler. This is a Tekmar and is controlled by the software...hard to get used to for troubleshooting.

But, things I've done: I changed the trap and the moisture trap. I put a sample heater on the sparger because it is mostly the heavier analytes that have the problem, so I thought that maybe they needed a little help purging. I did a leak check on the P&T using the check in the software and it came out fine. I manually injected my IS and Surr into the 40mL vials to bypass the standard injection on the Solatek 72. That's about it for the P&T side of it...

Where do I start checking out the P&T?

Thanks,

[JI2002]

You should try cleaning ion source and replacing the filaments as well.

[krshaffe]

JI2002,

Sorry I forgot to mention...Over a years time, I've cleaned the source about 10 times and replaced both filaments.

Thanks,

[Peter Apps]

Hi Katie

I am starting to get as puzzled as you are. Early on you confirmed that you were seeing positive deviations from linearity - in other words that the higher concentrations gave bigger peaks than expected. Then in your 11 March post you give peak areas that are SMALLER than expected for higher concentrations, and SMALLER in the presence of samples. This is easy to explain by detector overload (which is what the Agilent rep was telling you) but all of the other suggestions on the forum will fix positove deviations, not negative ones. Positive deviations are trickier to explain and more difficult to troubleshoot.

In your 5 March post you referred to direct injections being inconsistent. I thought that you meant injections of liquid samples direct to the GC inlet (in other words with the P&T disconnected). From your second to last post I now see that you were loading the standards etc into sample vials and then (presumably) running them through the P&T.

If you really do have positive deviations from linearity you need to isolate the various components of the system to localise the problem. Disconnect the P&T and make liquid injections (split) preferably with an autosampler, direct to the GC inlet, with the same quantities of each analyte going onto the column as you would have for your P&T calibration curve. If you still have the non-linear response the problem is in the GC-MS, if not it is in the P&T.

Over what range are you calibrating (in terms of mass of analyte on the column)?, and what are the compounds in the standard ?, is the strength of the effect related only to the concentration (mass), or are compounds of different chemical class affected differently ?

Peter

[krshaffe]

March 11 post: I'm using three internal standards and the last one is coming out the worst. 1,4-dichlorobenzene-d4. In my 10ppb standard the response is 153000 and in my 200ppb standard it is 247000. Then when my samples start it goes down to 98000 and remains pretty consistent.

If the above paragraph is what you are talking about...these numbers are supposed to be the same. They should not go up at all because this is an internal standard and the same amount is added to each sample/standard. What I have been describing are two different problems that I thought was brought on by the same thing. 1. I am seeing positive deviation from linearity in my analytes and 2. my IS and surrogate response is rising with increased standard concentration. I thought they were due to the same thing because both are only happening to the heavier analytes.

I did both...I did liquid injections directly into the GC that were inconsistent and I did injections directly into 40mL vials which came out the same as if the autosampler had injected them. In the last post I was only telling you what I had done to check the P&T system...I thought that was what you were asking. Sorry to confuse you.

[Peter Apps]

Hi Katie

Thanks for clarifying. You have to do this step by step. You have not answered the question in the last paragraph of my post, and now I also need to know:

when you did the injections to the GC inlet did you use an autosampler, or did you do it manually ?, and what solvent did you use, what was the inlet temperature, split ratio and oven temperature programme ?.

If you do multiple injections with either the P&T or direct to the inlet, what is the rsd of the peak areas under each condition?

If you plot raw peak areas of analytes do these also show positive deviations ?

Peter

[danicrd]

Dear Katie
how model of purge and trap are you using?
I'm ready to bet for Tekmar LSC3000 or 3100... right?

[chloesabrina]

Katie,
I found your post and noticed you hadn't posted in awhile. I am having the same problem as you and have tried multiple things to help to stabilize this last internal standard. Please let me know if you have any progress!!