Can't ionize a mystery compound - ESI and APCI on MSMS

[baldy]

Hello - I'm having a difficult time.

I have a mystery compound that apparently is a chromophore because I get a pretty strong absorbance in the UV - I'm using 228nm. I want to get some information about this mystery peak using my MSMS.

I've used my Sciex MSMS to try to elucidate anything about this compound - but I cannot seem to get it to ionize and read on the MSMS. Has anyone had experience with compounds that do not ionize easily? I've used ESI in positive and negative mode - APCI alike. Still nothing.

My current analytics are focused on raw plant extracts - so unfortunately this compound can be a lot of things. Any advice on how to go about resolving some information would be much appreciated.
Thanks in advance.

[Don_Hilton]

Can you get the UV spectrum so that you can get the lambda max - and can you determine anything about pH sensitivity of the compund (including spectral changes with pH)? And at what pH (or pHs) have you used to attempt to obtain mass spectra?

If you can get some information that would give a clue as to functionality, perhaps an ideal pH could be selected for ionization or the cmpound could be derivitized to give you mass spectral infomation.

And, what m/z range have you used in attempting to detect the compound?

[baldy]

Lambda max is around 210nm. I do not know anything about this compound, acidic nature included. I'm running an acidic eluent system with formic acid. I have not tried changing the pH yet, but it is something to consider on my end.

All I know is that it strongly absorbs UV energy.

I understand I'm shooting in the dark here. Thanks for the help.

[JMB]

Some additional info. would help us to help you, so

1) has the Unknown been isolated as a single component, or are you running LC/MS on a mixed extract ? please provide details of column type and mobile phase.

2) it has lambda max at ~210 nm, but practically everything has this; are there lower intensity maxima at higher wavelength ? any indication of conjugation ?
3) what is the extraction procedure ?

4) if reverse phase HPLC, what is RRT of unknown relative to any known compounds in the mixture ? trying to establish relative polarity (presence/absence of functional groups -OH, -COOH, -C=O, -NH2, etc)

5) as previously asked, what is m/z range ?

JMB

[baldy]

Thanks for your reply.

No - in my lab the compound has not been isolated, which is becoming an increasingly important detail.

The UV/Vis absorption is very featureless. There is a very small foot of absorption from 260-280, but this is very minimal. The UV/Vis absorption spectrum tell me very little.

The retention time of the unknown is why I'm so interested - and my class of compounds have COOH and OH functional groups. It eludes at a nearly identical time to an often regularly seen compound, but is easily differentiated by the UV/Vis spectral absorption differences. I'm using very simple reverse phase HPLC with MeOH/formic and H2O/formic in gradient mode. 0.1% formic in both.

I've scanned from 100-550. My m/z range for my analytics is 175-375 for all compounds. I'd be amazed if this "mystery" compound was much bigger or smaller regarding m/z.

I don't think there is a nitrogen on the compound. I'd also be amazed if it was a pure hydrocarbon that didn't have any oxygens. This compound is plant derived.

[Don_Hilton]

As requested above, tell about the isolation steps. You apparently have some reason to believe that there is no nitrogen in the compund - and the basis for that might help us. You say that your class of compunds has either OH or COOH - and I presume that you believe this compound to be a meber of the class. Again there has to be some information that you use to draw the conclusion.

And LC/MS is not my strong suite, but I'll ask: 0.1% Formic. If we are looking at something that is likely to be an acid, are we in the range that might give us ion supression (preferential ionization of formate over the unknown)? The number that comes to my mind is more like 1 to 0.1 mM for the acid.

[JMB]

[There is a very small foot of absorption from 260-280, but this is very minimal.]
[The retention time of the unknown is why I'm so interested - and my class of compounds have COOH and OH functional groups. It eludes at a nearly identical time to an often regularly seen compound, but is easily differentiated by the UV/Vis spectral absorption differences.]

Are we talking about polyphenolics here ?

Does the "often regularly seen compound" have a name/structure ?

Extraction procedure ?

JMB

[JMB]

Can we assume that OTHER components in the HPLC/UV chromatogram are ionized ?

If so, ESI and/or APCI and +ve and/or -ve ionization mode ?

JMB

[lmh]

100-550 is a very small mass range. If you're working on crude plant extracts (I believe you said you were?) then you will undoubtedly have loads of things outside this range.

Some things simply don't ionise. In some electrospray instruments, quinones are disastrous. In almost any LC ionisation, things without useful functional groups are very hard. Some things ionise, but not terribly well (sugars in electrospray, for example), so they can be present in quite high concentrations but still give you no signal.

If you can't get it to ionise, rather than spending a very great deal of time and effort trying, and ultimately perhaps getting a mass, but still no idea whether you've got the mass of the thing that absorbs(!) or what compound the mass corresponds to, there's a case for fraction-collection and sending it to someone who can do a more general technique such as NMR.

Beware of believing that an unknown peak is related to your compounds of interest without additional evidence: UV is only valuable if the wavelength maxima show a little specificity to the family of chemicals that interest you. Retention time is of very little value (everything has to elute somewhere, and unfortunately there will be many compounds with similar hydrophobicity but wildly different nature). It's best to keep a very open mind about the identity of a peak, unless you have strong evidence that it must be similar to something you know.

Seriously, consider fraction-collection and an alternative technique, at least to tell you you're looking at an organic compound and not something totally weird.

[paul_b]

There are some good suggestions on how to ionise difficult compounds here, but there is an alternative explaination - volatile species. These will often have limited UV at higher wavelengths but if sufficiently retained can give a normal UV peak; we typically see DCM, Toluene or chloroform exhibiting this behavior. If there is a GC-MS around you may be able to get some information.