GC/MS Systems, experiences and reviews

[panoramix1982]

Dear colleagues,
i 'm new in this forum and i would like your assistance-experience on GC/MS systems.
I'm responsible for procurement of a GC/MS system able to analyse (qualitatively and quantitatively) liquids resulting from biomass treatment.The liquid mixture is complex consisting of hundreds of compounds (water, alcohols,carboxylic acids,carbonyls,hydrocarbons,aromatics phenols,ethers esters,sugars,phenols etc.)

Wee have been offered the following GC/MS systems:

Agilent 7890GC/5975C MSD
Shimadzu QP2010 Ultra
Perkin Elmer Clarus SQ8C
Thermo Fisher Sientific ISQ

We want to have EI as well as CI modes available.
I would like you to share your experiences with me for the above mentioned systems (reliability,technical service, software etc).
If somebody has another instrument to suggest he/she is welcome!.
Thank you very much for your time and your help i really appreciate it!

[Don_Hilton]

For instruments to look at highly comples systems, dont forget to look at GCxGC-MS systems. I personally like the TOFMS systems but I believe there is a quadropole MS system out there that supposedly does this kind of work as well. For TOFMS, Leco and Zoex come to mind immediately. I believe that Schimadzu has a quadropole based systems.

When looking for instruments, ask the vendors to provide local references. There are several advantages to this.
1) These people know how good or poor the local support is for the instruments. This can vary widely for a given instrument manufacturer.
2) You can visit labs and get a feel for how easy/difficult it is to use instrument systems - and for the chemists giving the reccomendations. (There are some people who find a light switch to be difficult to operate...)
3) You may acquire a network of other users of the instrumentation, which gives you someone to talk with as you learn to use your system.
4) These people can tell you why they did not purchase the other sytems - and again you can evaluate the source of the advice as well.

If you have some particular problems where performance of the instrument is critical for success for your project, send the vendor some samples. You want to send samples that you know well, so you will know what the instrument can do. And give the vendor as much information as possible on how to perform the analysis -- you are testing the capability of the instrument, not the skills of the applications chemist.

[tlahren]

Don's advice is very good.

My personal $0.02 is that I'm sure all the vendors are near equal with the specs and performance of their systems. I have never operated a Shimadzu or PE system. I have a Thermo Trace GC and Agilent 6890 side by side and I always prefer the Agilent GC to the Thermo GC. I It may be due to my unfamiliarity with the Thermo system as every system has it's own "quirks." Like Don said, if you know some people (a network of other users) who all use one brand then it may benefit you to go with that as they will probably be of huge assistance to you later.

I find the Agilent 7890/5975C system to be an extremely sensitive system (when set-up and maintained correctly) and also extremely rugged and reliable. On the other hand I know the Thermo ISQ systems have better mass accuracy than the Agilent systems but it is still not considered "High-Res." I believe they have a no-vent style ion-volume removal system similar to my Thermo DFS. This is convenient in that you don't have to vent the analyzer region to clean the ion-volume or replace a filament. However, venting and cleaning is fairly trivial on a new Agilent 5975C in that you can vent, remove and clean a source, pump back down to working vacuum and be analyzing samples again by the end of the day if you are good at it. I also don't care for Thermo's Xcalibur software either. But probably because I'm used to Chemstation. I export all my Thermo Data and process it in Chemstation.

One more thing: The Agilent system says that they can do combined EI and CI on the same source and that you can run all modes (EI, PCI and NCI) in the same sequence. This is true only for crude testing and some limited research. If your sales rep doesn't downplay it's capabilities for you then I am trying to now. Just don't expect to be setting up a daily analysis method where you are doing both EI and CI together. It's also not reliable for quants. Again This is only true for the "Combine Mode." The ion sources used independently are rock solid.

As I said, my choice is biased and Don's advice is good. I hope you are able to have some fun shopping (I like spending my bosses money).

[Amirav]

You may also look at the Aviv Analytical 5975-SMB GC-MS with Cold EI.
http://www.avivanalytical.com/Supersonic-GC-MS.aspx
It provides molecular ions to all the hydrocarbons (unlike CI) and other compounds with a one sensitive Cold EI ion source, they have a molecule identifier software that provides elemental formulas and confirms or rejects NIST identification and it can analyze much larger biomass oil compounds than any standard GC-MS, up to C72 and including triglycerides (faster too). It can also analyze the fatty acids without derivatization. They also claim for having uniform response for better quantitation without compound based calibration.
You may ask them to demonstrate the above on your evaluation sample.
Amirav

[tlahren]

Amirav is right. I can't believe I forgot about this option. It is an incredible advancement in GC-MS technology. And if I'm not mistaken, it's capable of being interfaced with an LC for LC-EIMS as well.

Amirav, Do you know if Aviv Analytical sells the complete 5975-SMB systems or is it just the SMB interface?

Please correct me if I'm wrong on any details Amirav.

[Alexandre]

If you are after fine analysis/tuning of fermentation/process, e.g. to distinguish between Chimay Red and Chimay Bleu then you really need GCxGCxMS.
If you are after smell problems you better have MSMS as some borneols have 5 ppt smell threshold.
Sometimes to check that the biomass process is under control, the monitoring of markers (major relevant peaks) may be good enough. Then MS is not needed.

[panoramix1982]

Dear colleagues/friends,
thank you so much for your prompt replies and useful information.

Don, i have asked for a GCxGCxMS option but my worry is that some compounds are thermally unstable and extended time (2nd GC column)at high temperature might result in their decomposition.
The advice for asking for local references sounds nice.
(Actually only Shimadzu has offered me to pay a visit to a costumer that they have and uses their instrument.They also proposed me to send them my samples to test there.)

As far as the network of people is concerned, as you know Agilent is the most famous GC/MS instrument supplier and thus most of people are using their products(even if some of them are old models).

Tlahren thank you so much for the advice concerning EI and CI.In fact Agilent representatives told me that there is no compromise that i have to make when using EI CI on the same source and the accuracy is sky high even when performing quantitative analysis.

from a first glance qp-2010 ultra by Shimadzu looks a bit more advanced in terms of resolution ,scanning rate etc,however as i mentioned above almost everybody uses Agilent's products and that makes me dubious over the "superiority"(in fact the specs are more or less the same) of the Shimadzu instrument and more over the service that those companies provide as well as the easiness of the software are important factors.

Amirav thank you for one more option but i couldn't find any distributor/reseller in my country (Sweden).Also 3 months delivery time (as it is stated in the webpage) is too long for my boss.however if possible i would like to have more information about this...

Alexandre, the samples are actually liquids produced from biomass pyrolysis.

Once again thank you for your valuable comments and i'm looking forward to more of them !!!

[Don_Hilton]

In GCxGC the time spend in the second GC column is not long, nor does the temperature have to pe particularly high. I frequently use a configuration with the first column is a 20 m by 0.2 mm id column followed by a 1 m by 0.1 mm ID coolumn running at 5 or 10 degrees above the first dimension column.

If you have a sample, contact the GCxGC vendors and let them take a crack at the separation. Depending on your constraints and needs for confidentiality, this sounds like the kind of separation that a GCxGC vendor might like to showcase so this could be a customer demo and publication rolled into one!

[Amirav]

Panoramix1982
If as you wrote you would like to hae more information please write me at
amirav@tau.ac.il
Amirav

[panoramix1982]

[quote="tlahren"]
One more thing: The Agilent system says that they can do combined EI and CI on the same source and that you can run all modes (EI, PCI and NCI) in the same sequence. This is true only for crude testing and some limited research. If your sales rep doesn't downplay it's capabilities for you then I am trying to now. Just don't expect to be setting up a daily analysis method where you are doing both EI and CI together. It's also not reliable for quants. Again This is only true for the "Combine Mode." The ion sources used independently are rock solid.


Actually the Shimadzu seller referred to this issue and he actually mentioned that sensitivity is reduced in such a mode,however Agilent seller didn't mention anything about that, however he offered me an other instrument.
Do you think this a scale of "honesty" of the seller or is it just a result of a more flexible and broader range of Agilent products?

[emts]

Panoramix1982,
You may also look at Bruker SCION SQ:
http://www.scionhasarrived.com/

[Don_Hilton]

Noting the questions about a particular sales rep: Remember that you are puchasing the instrument not the rep. Don't hurt your self by staying away from a company that sells a good instrument unless the sales rep will somehow interfere in service or support in some way.

This is where sending samples to the potential vendor and getting references really help. The sales person may tell you that the instrument can do all the stuff you see on CSI. But with your sample, the instrument does no more than what the instrument does.

Waiting after an instrument is first introduced can be good. A vendor typically really believes in the new product -- and may have to work a bit to make the instrument actually do what they believe that it is doing as they introduce it. Or they may have to back off on claims after the instrument is actual use for a while. -- And I've seen this with many vedors who produce fine instruments and have an excellent reputation. And if an instrument has been in the field for a while, there are references -- which helps you. On the other hand, if the new instrument appears to have features that address a need you have, the vendor has a good reputation, and with your samples, you see good results, the new instrument may present low risk and high benefit to you.

I worked for an instrument company for a while and I remember a sales person who over promised on instruments. This particular person had a short career with the company. Typically a company does not like the guy who willfully oversells an instrument. There is a lot of cleaning up to do after the sale - and it costs the company reputation as well as money in providing application support to attempt to do what was promsied.

[HbJ]

Panoramix1982,
You may also look at Bruker SCION SQ:
http://www.scionhasarrived.com/

Yes, for misery and pain.

[splean]

Panoramix1982,
You may also look at Bruker SCION SQ:
http://www.scionhasarrived.com/

Yes, for misery and pain.

What is wrong with Bruker SCION? In terms of specs (sensitivity, vacuum system perfomance, ion source and pre-filter design) it looks quite interesting. Do you have any negative experience?

[tlahren]

I have nothing against other vendors. I just prefer the Agilent systems. In my opinion the Dual EI-CI capability is a cool idea but I never use it. You may be able to achieve some adequate quantitation using it. I was just saying not to expect to use it in place of the standard EI source for running regulatory type methods. I just wouldn't trust it to be that reliable. I would expect Agilent to use it as a sales point and try to say that the sensitivity is the same (under perfect Agilent lab conditions) to try and out-do other vendors. The best features of the Agilent are the new Triple-Axis detector (far better sensitivity than the previous multiplier system) and their long history of having rugged and reliable ion sources and quads. The Dual Sim-Scan mode is also awesome. But I would expect other vendors to have this option as well now that technology is jumping so fast.