Diff bet FID/GCMS, standards, inlet liners GW or no GW?

[new2GCMS]

Hi,
Can anyone help me with a few questions please?
I have an Agilent 6890/5975 with an FID on a separate port to use. I've been asked to quantify components in a couple of flavorings and I'm new to GCMS. Nothing in my book talks about what would affect my results.
I get dramatically different chromatograms from the Ms than from the FID. The FID is vastly simpler and the heights of the peaks are not the same.

I was curious to know how I would figure out if the instrument is running well. I figure I need to get standards of some type, but don't know what to use for an instrument that will be primarily studying organic acids and flavor compounds.
What do I use for standards? I saw that there are standard kits, but don't know which to get and the Agilent website doesn't give me a lot of info.

I also wanted to know if there is any difference that an old single taper inlet liner with glass wool (it was really dirty so I cleaned it with bleach and baking soda, then put a clean glass wool plug in it.) will have with the chromatogram from a new one.

I'm really lost because we didn't do a lot of GCMS in school and I'm new at this job.
Any help would be greatly appreciated.
thanks,
Eva

[Don_Hilton]

The question you ask covers a lot of territory. My first thought is that you need to ask someone with whom you work who has experience with the instrument. If the instrument is new, perhaps you can get some help from whoever helped to determine that the particular instrument you have is the one that meets the need.

As an alternative, take a class from one of the consumables suppliers on GC. This will help you to understand the selction of columns, liners, etc. and will give you and idea of how to find operating conditions that will help you to solve the problems you will encounter.

On what will affect results --- many things. Some off the top of my head:
Solvent for diluting flavorings.
Inlet maintanance. The septum and inlet liner need to be replaced with sufficient frequendy. On old septum will leak. A dirty inlet liner will "eat" compunds passing through it and will cause peak tailing.
Selection of an appropriate inlet liner and inlet termperature.
Selection of split or splitless mode for injection
Depending on mode: split ratio or purge off time.
Injection volume - 1 microlter works in most cases. Depending on the application, you may deviate from this.
Column age and condtion.
Selection of the appropriate column for the analytes of interst. For example, compounds with highly polar functionality (like glycerin, PG, and organic acids) tend to work better with wax type columns. Just about everything else in a flavoring does well on a 5% phenyl column. And for quantification, you may want to derivatize the polar molecules and chromatograph them on the 5% phenyl column.

On cleaning inlet liners - ther have been several discussions on the topic on this site.

On comparison of FID with total ion chormatogram (TIC): The FID tends to show a similar trace to the TIC, particulary as the TIC covers a wide mass range. Individual ion traces will vary greatly. The FID has a greater dynamic range, so the FID may give taller and sharper peaks for some of the large peaks seen in the TIC. And there are disffernces in sensitivity for some specific molecules. There has been some receint discussion of TIV vs. FID reciently on this site.

For knowing if the instrument is running well: There was some kind of checkout standard run when the instrument was installed. The checkout standard was specific to the detector in use. It probably had a specific chromatographic column specified as well. The checkou standard is a good starting place for checking instrument performance. YOu may be using a different column, in which case you will need to create a method which will give a chromatographic peak of similar width to that specified in the instrument test. You may not match the instrument installation specification with the particular instrument configuration you are using, but you have a measure of performance and you can watch to be sure that you are consistant.

On this site, you will see some discussions related to running flavors.

Good luck

[new2GCMS]

Thanks, so the instrument was purchased used. It has been run by my supervisor, but he isn't very familiar with the details of GCMS, so basically I don't have someone to ask.

I have just gotten new inlet liners and replaced both. I tried to clean them using info from discussions on this forum, then decided that I just couldn't be sure I was doing it well or right, and bought new ones. I don't know how to choose them, but I'm using single taper with GW for everything but SPME.

How would I find out how the inlet temperature will affect things? Is there a book you would recommend to get me up to speed?

I'm going from 50:1 split mode and tried pulsed split and regular split. I get nothing showing up in the chromatogram with the regular split. I have also tried splitless with 1 uL. In all three cases I get very little abundance 500000 or less counts.

Most of our flavor chemicals are nonpolar so I've an HP-1 column.

Thanks so much!
Eva

[Don_Hilton]

Single taper with GW should work for a wide variety of flavor samples. Deactivated liners and wool are best. For inlet termperature, you neet a termprature hot enough to move the compunds of interst onto the GC column but not destroy sensitive compunds. With a HP-1 column, I would start with an inlet temperature at about 275 degrees for most flavoring mixtures that I have encountered.

The use of split vs. splitless and selection of split ratio depends entirely on what concentration the analytes are in the mixture you are looking at.

If you are looking at mixtures of organic acids on an HP-1 column, you may have some difficulty unless you derivatize the acids.

In working with an unfamiliar technique or situation, I like to take a bottle of inexpensive perfume and run it. It will show a large number of peaks. Start with a low column temperature to start with. Inject and hold at that tempreature for a minute or so, then ramp the oven up at a moderate rate, like 3 to 5 degrees per minute and look to see what comes off. Be sure to keep the filaments off during the solvent peak. Collect in full scan mode - 45 - 350 mass units and see what you have.

The nice thing about working with the perfume sample is that you don't worry about gettign a wrong answer, you just begin to tweak conditions and judge the chromatography.

[Don_Hilton]

And regarding a book. A lot of the practical how to that I have learned is from working with others who looked at what I was doing and had pointed out possible improvements in my conditions. There are probably some books out there, but for the day-to-day set up of methods and operation of the instrument it is like driving a car or playing a piano. Mentoring helps. For selection of liners, columns, and conditions browse the chromatorgrams provided by GC consumables suppliers and brows applications notes.

[gstaepels]

New2GCMS,

I have an interesting document that might help you chosing the right injection liner. Contact me using email gpstaepels at mmm dot com and I send you the file.

Gilbert

[new2GCMS]

Don,
The issue I have with getting a mentor is that there is no one in my division doing GCMS, so I'm not in a position to get much of other people looking over my stuff and telling me advice on what isn't working. thanks for the advice though.

Gilbert,
I've emailed you offline. Thanks!

[JonKlar]

Do you know what analytes you are looking for or are they asking you to identify and quantify unknown components?

[cody84]

Definately not a pro at GC but I have attended some varian training. And I haven't done MS so bear with me:

In GC I think the idea is to volatilize your whole sample (~1 uL injected) in the injector and then have it condense or 'focus' on the inlet of the cooler column. FID is supposed to be something like 10-20 degrees hotter than injector. The column oven should be set near the analyte's boiling point and then you can slowly ramp the temp up and see how the chromatogram looks. Once you set the injector and detector temps, then you can focus on playing with the oven ramp and flow rates etc until you have nice peaks.
Hope that helps...but I'm not all that confident I don't have much development work on GC...just HPLC.
As for MS I have no idea. Sorry!

[new2GCMS]

JonKlar,
They want me to analyze unknowns.

Others:
one problem I found out is that the people who refurbished the instrument installed the wrong FID jet and had a bad nut and ferrule there as well. Apparently the 6890 uses a shorter jet than the 7890 and the 5890, so that's one reason the FID data looked so different, is that the longer jet and the bad connection meant that I wasn't getting the flow to move properly to the detector.

The other thing I have learned about the system is that on the mass spec side the ion source was dirty and needed to be cleaned and the multiplier when installed was seated improperly. so it gave erratic results.

Ongoing problems:
Despite all this, I still can't get reproducibility within runs. What troubles me the most is that in a mixture of flavor components I get varying RSDs for each component. t he percentages are anywhere from 2 to 48% for an individual component within the overall flavor say for example something like orange. If I run the same orange sample liquid injection seven times, and then look at each peak, the separate components that I know make up the flavor since we made it in house don't have the same peak areas. Any thoughts as to why this might be?

[Don_Hilton]

The question on reproducability in components in orange was posted either under "Gas Chromatography" - and there is a reply over there...

[Peter Apps]

When you say that you have GC and FID on separate ports, does that mean that you have two inlets and two columns ?, or are you switching one column between inlets ? Or do you perhaps have one inlet and column with a split between the FID and MS at the end of the column ? Or something else ?

How are you doing your injections ? autosampler or manual ?

Peter

[new2GCMS]

@Don_Hilton
It isn't just orange. This happens with blackberry, cherry, even with a mixture of 50% ethyl butyrate 50% ethyl acetate. Is that still the same issue?

@Peter
Yes, 2 inlets and 2 columns, both columns are Hp-1. injections are done with an autosampler --a Gerstel MPS-2

[new2GCMS]

I can't get reproducibility within runs of the same sample. What troubles me the most is that in a mixture of flavor components I get varying RSDs for each component. The percentages are anywhere from 2 to 48% for an individual component within the overall flavor say for example something like orange. If I run the same orange sample liquid injection seven times, and then look at each peak, the separate components that I know make up the flavor since we made it in house don't have the same peak areas. For example, the orange has ethyl butyrate in it, ethyl propionate, and maltol among other things. My RSDs for peak area for ethyl butyrate is 38%, for ethyl propionate 3%, and for maltol is 9%. Why would I get such variation in peak areas from one run to the next of the exact same sample. I have tried it where I have the same vial for each run and where I make up separate vials so there isn't an issue with lowered volume affecting draw up in syringe with the autosampler (Gerstel MPS-2). I get the same issue regardless of how I handle the vials.

the operating conditions are:
inlet: He carrier gas, split injection 50:1 ratio, 54 mL/min total flow
(tried also 20:1 split ratio 24 mL/min total flow
and
100:1 split ratio 104mL/min total flow --neither of these were any better with the RSDs either)
19 psi in inlet @260 degC with a low PSI drop universal ultra-inert liner with glass wool.
column: capillary HP-1, constant flow, 19psi, 1mL/min flow, 26 cm.sec avg velocity
oven: 70degC, ramp is conservative with every 5 min adjustment by a few degrees.

Any thoughts as to why I'm getting strange RSDs for individual components and how I might troubleshoot it?

Thanks,
Eva


_________________________________________________________________________________
Don_Hilton
Post subject: Re: poor reproducibility with peak areas of flavors
Posted: Mon Jun 11, 2012 3:58 pm

Are your injections with an autosampler or manual?
How often to you change the inlet liner and the septum?
How well are the components separated from each other - and other stuff present?
Can you post a picture of the chromatogram?
And what kind of detector? FID? MS?
_____________________________________________________________________________
Post subject: Re: poor reproducibility with peak areas of flavors
New postPosted: Mon Jun 11, 2012 11:03 pm


What solvent are your flavour compounds dissolved in ?

Having the same problem in two forums is going to get very confusing - if you and don are agreeable, let's put it all in the GC-MS thread

Peter

_________________
Peter Apps

[new2GCMS]

Don--
The inlet liner gets changed roughly every month or two. Because I also do SPME in headspace and headspace needle injections for top notes and colored compounds and only have one instrument, I often change the liners in and out (dedicated liner for SPME requires this), so when the liner looks fouled -color or deposits on the glass wool, I dump it and use a new one.

The septum gets changed every 2 to 4 weeks.

For now, I am just doing analysis on the MS side, because only last week did I figure out that the FID jet was the wrong length for a 6890GC so I'm not up to speed yet with the FID.

I haven't been able to get a chromatogram to upload on the website yet, so as soon as I can get that to work I'll post it.

Peter--
solvent is ethanol usually or
ethanol and glycerin or
ethanol and propylene glycol
less commonly ethanol, propylene glycol, and glycerin

I've consolidated everything over here now so it will be less confusing.

Thanks,
Eva