Looking for advice on our Thermo Dionex Ultimate 3000

Discussions about GC-MS, LC-MS, LC-FTIR, and other "coupled" analytical techniques.

5 posts Page 1 of 1
Summary: We purchased an LCMS system second hand and have spent almost 1 year trying to get it working for our applications. I am looking for impartial advice in order to get a fully working system.

-------------------

I own a small research laboratory in the Netherlands. We offer custom synthesis services and therefore need to be able to qualitatively analyse reaction mixtures as well as verifying the purity of finished products. We also receive some external samples from customers that require analysis in order to identify the product and confirm purity (they usually provide us with the target molecule structure so we know what we're looking for).

We have a Dionex UltitMate 3000 LC System with DAD and Thermo TSQ Access LC/MS/MS with HESI II Source. We set the system up ourselves and have the LC component working very nicely but the MS component is nothing but trouble - the masses are always way off. After spending months chasing Thermo (their support is terrible) we managed to get an engineer out and he changed some settings and determined that one of the quadrupoles was faulty (Q1 I think). He said that it is fine though and will only affect the masses of higher molecular weight compounds (>900 Da). That is OK for us since we are working mostly with small molecule drugs right now.

However, the issue isn't fixed. The masses are still way off (even for small molecules) and there is often a lot of background noise in the spectra. The only way to get accurate masses in the spectra is to inject a reference sample directly into the MS and add this data to the tune file. Of course, this isn't possible when you're trying to identify novel compounds that you don't have a reference for.

Another problem we've noticed is that the sheath gas never goes above 15 units (it should be 35 as default). The nitrogen generator is new and is able to deliver 30 L/min at 100 psi. Thermo thinks it might be a valve in the MS. Could this be contributing to our problems?

Due to the ongoing problems I have been contacting other suppliers of MS systems to see if we can swap out our TSQ for something that actually works. Agilent seems to think that the TSQ isn't suitable for our needs. They say this is a 'quantitative MS' and what we need is a 'qualitative MS'. They are suggesting a Q-TOF.

Of course, Agilent (and any other supplier) wants the sale so I cannot trust their recommendations. This is why I am posting on this forum, to get impartial advice on this issue from the very knowledgeable people here. As you can probably tell, I am not an analytical chemist. I have an organic chemistry background but I have not spent much time with the hardware side of analytical chemistry.

Any help would be much appreciated! Sorry for the book.
Hi & welcome to the forum.

While I don't know in detail what you are trying to do with the instrument, I think Agilent might have a point. Qualitative analysis of reaction mixtures is not the strength of an LC-triple quad-MS. (A combination of) other techniques are better suited for answering the question "which molecules are in this mixture?", such as LC-Q-TOF, or NMR for structure determination. If what you are trying analyse is semi-volatile, a much cheaper alternative is GC-MS, allowing database searches/comparison.

Purity determination can be done by LC-triple quad-MS, but I would prefer LC-UV. It's cheap(er) and robust.

About the masses that are off. This should have nothing to do with the sheat gas pressure. From what i read, the instrument is not properly calibrated. Can you give an example of the mass offset? Are you sure it's not an adduct you're seeing?

It is not surprising to me that Thermo (or any supplier) isn't keen on supporting systems that were not bought from them. I suggest to look for third-party support in your case. I'm from a Belgian lab, and I can recommend MSVision for support (they're from the UK and can work well with old Thermo systems, we have a couple of those).
Thanks for your help.

It's not like we need to do complex structure elucidation. I mean when carrying out a reaction we know what product we want so it's just a case of spotting the target molecular ion or a fragmentation peak thereof. What we want (in most cases) is a quick and easy way of answering "did this reaction work". I don't see why this TSQ can't do this.

Of course, it would be great if we could identify unknown mixtures and reaction side-products etc but at the moment we're just trying to get somethig moderately functional working.

As mentioned, the LC component works quite well and is coupled with a DAD so we can determine purity but that assumes we characterised the molecule by some other means.

We do have GCMS but it is not suitable in many cases. We work with some compounds that are quite heat sensitive and not very volatile.

Regarding the masses, let me give an example. One molecule we were looking at has a mass of 323 and we were getting anything between 317 and 320. This cannot be an adduct. The masses are always less than they should be (unless we first add all the daughter ions to the tune file by directly injecting the reference into the MS). Perhaps it's not properly calibrated but then why can it give accurate spectra when it is analysing a compound in it's tune file? Surely there should be a way to calibrate it so that there is a general method for determining unknown samples? We've calibrated using polytyrosine numerous times too.

What I also don't like about Thermo is their software (Xcalibur). It is very difficult to use and doesn't have the features I need. Do you recommend any software in particular? What would be great is something simple like Agilent's GCMS software, with the ability to search a NIST database.

Thanks for the MSVision suggestion, I will contact them.
synex wrote:
Thanks for your help.

It's not like we need to do complex structure elucidation. I mean when carrying out a reaction we know what product we want so it's just a case of spotting the target molecular ion or a fragmentation peak thereof. What we want (in most cases) is a quick and easy way of answering "did this reaction work". I don't see why this TSQ can't do this.


The instrument should indeed be able to do that.

synex wrote:
As mentioned, the LC component works quite well and is coupled with a DAD so we can determine purity but that assumes we characterised the molecule by some other means


Sorry, I missed that in the first post. Since you have the LC-DAD working well, could you use that as a (temporary) analysis tool to quantitate a known reaction product or leftover reactant in a reaction mixture? The retention time and DAD spectrum are also characteristic for the molecule. It depends if you have a good LC separation and if the molecules in question absorb in the DAD region you're measuring.

synex wrote:
Regarding the masses, let me give an example. One molecule we were looking at has a mass of 323 and we were getting anything between 317 and 320. This cannot be an adduct. The masses are always less than they should be (unless we first add all the daughter ions to the tune file by directly injecting the reference into the MS). Perhaps it's not properly calibrated but then why can it give accurate spectra when it is analysing a compound in it's tune file? Surely there should be a way to calibrate it so that there is a general method for determining unknown samples? We've calibrated using polytyrosine numerous times too.


It definitely sounds like there's an MS problem that needs an on-site fix. We gave up on trying to calibrate our old TSQ ourselves and let MSVision do it during yearly PM. For example, we were struggling with the software resetting the calibration for unknown reasons. I don't know if you have a similar problem and I can't tell if it's software or hardware related...

synex wrote:
What I also don't like about Thermo is their software (Xcalibur). It is very difficult to use and doesn't have the features I need. Do you recommend any software in particular? What would be great is something simple like Agilent's GCMS software, with the ability to search a NIST database.


You will have a hard time using software from supplier X to analyse data measured on an instrument from supplier Y. It's unfortunate, but it's the way it works: instrument manufacturers make their instrument data work with their software only. I personally think Xcalibur is OK for quantitative analysis, but I can see it lacking features for qualitative work (spectra comparison etc.). As far as I know, there are no (NIST) databases for LC-QqQ full scan MS data because ionization and fragmentation (if any) is not straightforward/reproducible like in GC-EI-MS.
Thanks for further details. Also thanks for suggesting MSVision, they have been very helpful so far and have already run some tests via TeamViewer. The issue is still there so I think we're going to get them over to clean the quads and look at things further.

Cheers :)
5 posts Page 1 of 1

Who is online

In total there are 11 users online :: 0 registered, 0 hidden and 11 guests (based on users active over the past 5 minutes)
Most users ever online was 599 on Tue Sep 18, 2018 9:27 am

Users browsing this forum: No registered users and 11 guests

Latest Blog Posts from Separation Science

Separation Science offers free learning from the experts covering methods, applications, webinars, eSeminars, videos, tutorials for users of liquid chromatography, gas chromatography, mass spectrometry, sample preparation and related analytical techniques.

Subscribe to our eNewsletter with daily, weekly or monthly updates: Food, Environmental, (Bio)Pharmaceutical, Bioclinical, Liquid Chromatography, Gas Chromatography and Mass Spectrometry.

Liquid Chromatography

Gas Chromatography

Mass Spectrometry