_edited.jpg)
In my last post, I discussed the many benefits of using GC-MS in Extractable and Leachable Analysis. This time I want to turn my attention to LC-MS. LC-MS is GC-MS flashy younger sibling. It may have not been around as long as GC-MS but it is now starting to be an indispensable tool in the analysis of E&L.
LC-MS, like GC-MS has many factors which will influence whether a substance in your sample will actually be detected, identified and quantified using LC-MS. In this article, I will only focus on the major ones.
Unlike GC-MS, liquid chromatograph does not require the sample to be in the gaseous state, but it does require the substance to be soluble in the solvents you select for sample preparation.
If they are not soluble in the solvent used the substance are likely not be injected onto the chromatograph or filtered out prior to any chromatography. They also need solubility in the mobile phase to progress down the column and for course the nature and relative affinity of the substance for the stationary phase will determine how the substance elutes (or not) from the column and into the mass spectrometer. So, thinking about how you dissolve your samples and which solvents you choose is very important.
When it comes to trace analysis generally, the quality of the solvents you use is a key concern. You stand very little chance of identification and quantification of trace level substances in solvents which are not fit for this purpose.
So why choose LC-MS? Why not LC with another detector? The reasons for choosing LC-MS over LC with detectors such as UV are essentially the same as with GC-MS, namely high specificity and the ability to detect a wide range of complex mixtures at low concentrations.
Once it arrives at the MS once again there are many mechanisms which it might be subjected to, not least the process by which the liquid stream is introduced into the MS. In LC-MS there are two major types of Ionisation sources; Atmospheric Pressure Chemical Ionisation (APCI) and Electrospray (ESI) and variants around these such as APPI.
These sources both aim to remove the solvents used in chromatography and retain and pass on the substances of interest to the Mass Analyser. Different manufacturers sources may produce different outcomes and it is difficult to say any one source will ionise all substances which might elute from the LC analysis and there we have our major issue with LC-MS it is not a universal detector with a consistent output.
Whereas GCMS analysis from one manufacturer will pretty much give you the same result as from another, the same really cannot be said for LCMS. The source designs vary a lot, I don’t believe it is possible to expect the same result from different manufacturers systems because of this.
This gives a problem, there are many different manufacturers of LCMS unless you are prepared to run each sample through each manufacturer’s equipment you have to make a choice. Whatever choice you make that source / detector will have specific characteristics.
It is certainly possible to get consistent results but its far less certain between different sources from different manufacturers.
Unlike GCMS, there is no guarantee that substances which elute from the LC analysis will be ionised and consequently detected, you should consider LCMS a selective detector. The modes by which APCI and ESI operate are quite different and you will see different substances dependant on the source design and whether you choose positive or negative. This is before the substances are passed to the mass analyser which again like GCMS detectors can be a number of different types.
But what you will generally see with LCMS sources is a soft ionisation (compared to the EI source commonly used in GCMS). This means the ions passed to the mass analyser are more likely to be representative of the molecular mass of the substance. Having a ion representative of the molecule mass aids in identification and allows for other MS/MS experiments to give insight into the sub-structure of the molecule but identification of unknowns does not benefit from access to large spectral libraries of EI spectra which are possible due to the reliable and consistent spectra which a EI source produces.
However, the combination of a molecular ion (with its isotope cluster) and the fragments obtained from MS/MS experiments generate a data rich environment. It is now possible to take that data and search against large public databases such as Pubchem (97 Million compounds) and Chemspider (78 million structures) of molecular formulas.
That can be a daunting numbers and it very much depends on quality of the MS data as to how successful you might be. FM Greens paper in Anal. Chem 83, 3237-3243 (2011)
Showing that a 1ppm or a 10ppm mass accuracy makes an order of magnitude different in numbers of structures which might align with entries in database whilst these seem like huge numbers this can be quite effectively reduced with a small amount of extra orthogonal information to eliminate many potential structures.
So, can you make a choice?
Well let’s see.
GCMS universal detector response (almost) but selective chromatography – must be volatile.
LCMS wide range of analytes will elute from chromatography but lack of universal detection.
I would say it is like choosing your favourite child, impossible and not a smart idea to try.
The best you can do until the Star Trek tricorder is available (some claimed DESI was getting close - I would have to disagree for now) is to fully understand the limitation of each technique and be prepared to supplement each with other things such as knowledge of the material you are testing and other detectors and techniques which might provide assurance that you have not missed something.
No one technique is the answer, only a fully rounded approach and science led design can build up the pieces in the jigsaw to reveal the full picture of leachables present in your drug substance or extractables in your material.
Hope you enjoyed this blog. Please comment and/or pass on ideas for future articles.
You can Email me via Jason.Creasey@MavenEandL.com or visit the website www.MavenEandL.com
Next Time… What would people most like to see in a future ICH document on E&L?