Possibility for mass spectrometry to go wrong?

[Genecks]

Let's say I take a substance and put it in a mass spectrometer. I don't know what the substance is, so I want to know about its mass. Well, is there any chance something could go horribly wrong, such as the compound exploding?

 

What procedures do forensic scientists undergo before putting an unknown substance into a mass spectrometer?

[RyanJ]

I've never heard of that happening. Basically the process for entering an unknown substance into a mass spectrometer is as follows:

 

1. Load a sample into the MS instrument, and vaporize it.
   
2. Ionized the unknown substance (which results in the formation of positively charged ions)
   
3. Accelerated the ions using a magnetic field.
   
4. Computation of the mass-to-charge ratio of the particles based on the details of motion of the ions as they pass through electromagnetic fields.
   
5. Detection of the ions, which in step 4 were sorted according to m/z.

[Genecks]

During steps 1 and 2, I could imagine an energetic material being ignited/excited and thus undergoing an exothermic reaction. That's what I'm thinking.

[RyanJ]

There are plenty of tests that can be done before mass spectrometry to identify potentially dangerous compounds before they are analyzed. Otherwise you would be correct, step 1 in particular could be hazardous with certain materials.

[UC]

MS uses incredibly tiny amounts of the substances involved, so there's no issue. The worst that happens is that the sample decomposes or fragments.

 

The ionization can occur by a large number of processes, some of which are very gentle to the molecules (Things like MALDI work on giant proteins without destroying them). The most common is probably electron impact, which tends to fragment the parent ion to a large extent, allowing structural information to be determined.

 

It is possible to run a negative ion MS, but that requires atoms like halogens to be present.

 

m/z analyzers vary from standard quadrupoles, magnetic sector analyzers, time-of-flight (TOF) tubes, and the holy grail being ion cyclotron resonance. Quadroples are quick-and-dirty and work well enough on small molecules.

[CharonY]

The flow is too low and the vaporization to mild to allow for anything like that to happen. In LCMS, for instance you pump in and vaporize at most 1-2ml/min of your solvent (which is often flammable like methanol), without anything happening. And then you add a minute amount of your sample. The most you could do is mess up the ion source (usually by using unfavorable conditions and way too much substrate).

And just btw. depending on substrate and ionization you can also easily yield negative ions. You can run any MS in negative mode, which basically just requires a switch of polarity.

---

Merged post follows:

Consecutive posts merged

Just to add, apparently someone did manage to mess up his MS in a liquid system. Apparently a non-standard mobile phase (not further specified) was used that resulted upon heating into a volatile reaction. Of course the volume were too low to blow anything up but apparently the resulting products were covering the whole ion source. But again, it was not the analyte but the mobile phase.

Edited November 2, 2009 by CharonY