High 233/235 background with Methane NCI

[benhutcherson]

I'm a complete newbie to doing any kind of NCI(I've only done some PCI in the past, but not for a while) and unfortunately it's difficult to find a lot of published spectral data and so forth for any kind of NCI method.

I'm trying to find a few drug metabolites in samples that I know where they should be present, and have papers where I'm following the method exactly. I have standards for some-not all-and have papers for some compounds of interest where full NCI spectra are published and others where just the SIM peaks to monitor are IDed. Because I'm looking for a couple of things, I'm generally operating in full scan of 190-480-which covers everything-but I've tried SIM just to find the peaks I'm hunting for.

The biggest issue is that I'm seemingly not seeing anything even when I run standards. What signals I do get are swamped by a huge 233 and 235 background(233 is a little lower than 235, but not by much).

For reference, I'm using a Varian 300-MS triple, although operated in single mode. I have the source set at 200º, and have methane with a source pressure of 1.5 torr(filament 70EV, 50µA). It's tuned on PFTBA. I'm running hydrogen at 1.3mL/min through a 50mx.15mm SGE BP5(standard 5% phenyl methysiloxane) column, which I know is a bit of a beast but it's what's in the instrument now and has worked for me for several other samples.

Does anyone know specifically where this 233/235 could be coming from? I'm afraid I'm losing my analytes of interest for it.

Also, as a side note-in Chemstation I routinely "extract" ion chromatograms from a full scan TIC-i.e. if I have a chromatogram loaded I can tell it to show me a chromatogram for just 343, for example. I've yet to find any capability to do so-both in documentation and in just playing around-in MS Workstation 6.9.3. Does anyone know if this capability exists, and if so what I'm overlooking?

[JMB]

Ben,

I think that you have the two isotopic signals for Rhenium Trioxide (ReO3)- .

Re 187 @ 62.6%
Re 185 @ 37.4%

This implies that you have a fair amount of air in the source; possibly a leak.
If this is the case, the filament (rhenium) may burn out quite quickly!!!

Regards,
JMB

[benhutcherson]

Thank you JMB.

When I run an EI autotune, the instrument reports air levels as okay. I don't have a tune report in front of me, but the 32/28 ratio was correct with the absolute 28 level I think at 18(not sure what the scale is, but I know the report claims 20.0 is considered the max). Also, without the CI gas flowing, the source pressure registers ~.9mTorr and the manifold pressure is in the low 10^-7 range.

With that said, after revisiting it this week, I suspect that the issue is that there just wasn't anything of what I thought I was trying to measure there!

Also, as of today we're on a 3-week work from home shutdown. I'm going in Monday to shut down all the GC-MSs, so that will be the last chance I have to do anything for a while.

[JMB]

Ben,

When you say "shutting down all the GC/MS systems" what does that mean?

If it were me, I would leave ALL pumps (rotary/turbo/diffusion) ON and heaters on the MS ON; just turn the GC oven heaters to OFF.
In my experience, venting the systems and walking away for 3 weeks is just asking for problems.
Also, you should probably shut-down the H2 flow to conserve it.

Good Luck,

JMB

[aldehyde]

I agree, venting the instrument and leaving it off can cause more problems than leaving it idle. Hydrogen isn't a bad idea (and is cheap), nitrogen will also work. Just dont run the samples in nitrogen and make sure to give it time to purge back to helium before running or tuning.

In my experience you don't see much spectra with NCI, that's the point. It is a very soft ionization so you will primarily see molecular ion. If you had a TQ you could do product ion scan in order to SIM for the molecular ion of interest and then fragment it in the collision cell to get spectra.. but on a SQ it is mostly retention time + primary ions for detection. The benefit of NCI is decreased fragmentation + not ionizing matrix background (if using ammonia.)

[benhutcherson]

I haven't decided yet what to do.

I'm in agreement that GC-MSs are happiest under vacuum.

I have 3 that I look after now-an HP 5890/5971, an Agilent 7820/5975, and this 3800/300. I also intermittently have been trying to get a 3900/2100D going again, but it's not currently under vacuum. If it were, I'd be really hesitant to shut it down as one of my big issues is that it seems to take forever to get all the water out.

The 5971 is actually an easy one, even though I know that's not what I'm discussing here. I'm supposed to get delivery tomorrow of a whole big pile of 5971/5972 parts including the complete eMod and a vacuum manifold+CI manifold out of a 5972. Because of that, I'm probably actually going to haul the whole mess home with me and do my planned 5971-5972 field upgrade at home. Maybe that's ambitious, but a scientific instrument specialist needs something to do when not actually able to be at work .

Although I've known of systems that have operated for years(and rarely being opened at that) with carrier only flowing when the instrument is actively under operation, I'm uncomfortable with that. In particular, I worry about ingesting air through the column into the mass spec, and also the part of the column inside the transfer line getting wrecked. I have my HP/Agilents set up so that I can flip between helium and nitrogen, and run nitrogen in periods of inactivity. I get UHP nitrogen for about $17/cylinder from our gas supplier, so that's not a big deal. I don't have the 3800/300 set up for that(it's in a different room) but could add it on Monday.

One of my big question marks now is that I don't know if the current "3 weeks" will actually be longer than 3 weeks. When the decision came down today, I managed to negotiate clearance for 1 day/week to come in and check on things.

[benhutcherson]

If you had a TQ you could do product ion scan in order to SIM for the molecular ion of interest and then fragment it in the collision cell to get spectra.. but on a SQ it is mostly retention time + primary ions for detection. The benefit of NCI is decreased fragmentation + not ionizing matrix background (if using ammonia.)

This IS a TQ, so I could do that, and in fact I've done a fair bit the past week with PCI and product ion scan.

For whatever reason, though, I'm just not seeing the peaks for compounds for which I have either a literature published NCI complete spectrum(clonazepam) or literature describing SIM for the major metabolite(7-Aminoclonazepam). I have standards for these compounds, but can't see them in NCI on the standard. I can see them on PCI and have done so both to confirm their presence and get RTs-they're just not showing on NCI.

[aldehyde]

What CI gas are you using, Methane?

What is the ion source temperature and CI gas flow rate?

[benhutcherson]

Yes, methane.

Source temperature is at 200º. The Varian does not allow you to set a CI gas flow, but instead you set a source pressure. 3 torr is presumably the "default" since it complains when you autotune if it's not there. I've tried this running between 1.5 and 3.0 torr.

Not sure if I mentioned this also, but yes I do have a CI ion volume installed.

[aldehyde]

What pressure are you using for PCI?

I don't want to give too much advice because I've never used CI on a varian GCMS. The source temperature seems reasonable and you're seeing the peak in PCI so I would expect the GC method parameters and the MS in general are operating fine.

The high abundance of ions from a leak is concerning but even on Agilent GCMS NCI can be really hard on the filament.

What emission current are you using? What did you use for PCI and what did you see on the tune report?

The Agilent instruments I've used typically use double the methane in NCI mode v. PCI but I don't know if the same conditions would be applicable to a Varian gcms. Do you run autotune for the different modes w/ Varian or is it just a method parameter?

[benhutcherson]

I've been using 1.50 torr also for NCI. Should I increase it?

The emission current is at 50µA, which is the default setting.

When you run autotune on this instrument in CI mode, it gives you the option of doing NCI, PCI, and tuning on both Q1 and Q3. To give me the most options, after installing the CI volume, every autotune I've done has been on both quads and in both PCI and NCI.

I can grab a tune report and upload it when I go in on Monday. I don't remember the specifics of it, but needless to say PCI and NCI tunes for Q1 and Q3 turn into a several page long affair. It's worth mentioning also that Varians tune on PFTBA(or FC-43 as all their documentation/software says) for CI, where I know Agilents generally tune on PFTDT(or at least what my reading tells me-I've never used an Agilent with CI).

[aldehyde]

I'd be interested to see the reports. The filament current and gas pressure MAY need to be increased. I would check if the manual has any recommendations before changing settings.

[benhutcherson]

Thanks-I'll do that, and maybe do some playing around in what time I do have on Monday.

Unfortunately, at least no manual that I've found has really any guidance on this kind of stuff.

I'm use to-for example-the manual for the HP5971 where-for example-gives an overview explanation of what every parameter does and how changing it affects things like peak width and/or intensity. I've not seen that level of detail in any Varian manual.

I honestly have a bit of a love-hate relationship with this instrument-on one had it it forces you to get relatively hands on with everything and prompts you to adjust variables that you never event think about on an Agilent. As an example, for SIM(or product ion in a QQQ) I often like to decrease peak resolution a bit. On Agilents, the only way I know to do this is to create a custom tune file and append that to the method. On the Varian, in the MS method there is a dropdown box to set this for Q1 and Q3-it defaults to "calibrated" which uses the value from the last tune(generally ~.5), but you can narrow it down to I think .2 or widen up to crazy high values-I think 10 or maybe even more. Changing emission current, voltage, temperature, and even CI pressure is even simpler-rather than being tied to a specific method it's tied to the current instrument settings. From the nice little graphic that shows the instrument layout, you click on box next to the picture of the source and it brings up a window that lets you change all these parameters, and they do not change with the method-it just fusses at you(gives you a box where you have to check "okay") if you choose something outside the normal default.

The other side of this is that the documentation will show you where to go to change parameters, but doesn't really give you a lot of guidance about what effect changing them has(the GC manuals do this, but not the MS manuals). It seems to just assume you know it. I had never actually used a QQQ in my life before getting this instrument, and simple things like collision energy stumped me. You can prompt the instrument to do a "breakdown scan" where it scans the collision energy and plots the amount of fragmentation vs. CE. With a GC, however, it's only really practical to do this with PFTBA with as long as it takes(the basic chassis is very versatile, and could be factory configured for GC, LC, or with interchangeable sources to do either, and could even be filed upgraded to a TQ if ordered as a single).