Open Access LCMS: Strategies for trouble-free operation - PART 2
In Part 1 of this article I outlined how a vicious cycle of events can lead to contamination of an open access LCMS system such that it can become unfit for final compound QC leading to problems with compound registration and productivity. Further this was traced to the analysis of far too highly concentrated samples.
Here in Part 2 I present strategies to prevent these issues arising.
Sample Concentration Recommendations
As previously discussed the correct sample amount would be that which gives a peak height of the product equal to the upper limit of detection linearity so as to allow good detection and integration of relevant impurity peaks without over-estimating them compared to the product. Obviously we can make anything look pure if we only inject just enough to integrate a single peak – but that's not the object here!
So, what is an acceptable concentration? This will vary somewhat between compounds (chromophores) and the actual LC system and analytical parameters so some trial and error will be required but for a typical fast generic LC system (e.g. using 50 x 2.1mm id columns) this will be of the order of 0.25 to 0.5mg/ml for a 1ul injection.
This is easy to control for a solid sample but clearly much more difficult when sampling from a reaction mixture. However, if you know the amounts of reagents that went into it and the volume then it must be possible to make an estimate and, therefore, to take an accurate volume with a suitable pipette and to dilute appropriately. This sort of attention to detail is not particularly onerous and, once it becomes the norm, will go a long way towards avoiding the extreme error cascade situation.
Hopefully it goes without saying that samples should be inspected visually for any undissolved solids and suspect samples filtered, e.g. using filter vials such as are available from Whatman and Thomson.
Run regular blanks
In addition to running a test mix or system suitability sample first thing in the morning, preferably under automation, consider running blanks interspersed through the day. These can be integrated and, in the event of finding peaks above a certain user-set threshold, if desired the system can be paused pending intervention by the Analytical Team. Such intervention may be flushing the system with strong mobile phase, repeat injections of clean solvent to address contamination originating from the injection system or changing contaminated frits (pre-column filters), guard or analytical column as is necessary to remove the contamination. At least 2 blanks should be run consecutively and actions based on the results of the final one.
Take the pressure off
A strategy to reduce the need to run all reaction mixture samples directly on open access LCMS systems could include implementation of a number of measures including those suggested below. The benefit of all the following approaches is that, by running samples on these systems rather than the open access LCMS, the overall sample load on the latter is reduced and consequently they are kept cleaner for longer with better uptime and better results for critical QC analysis.
Siting an LC-UV in the synthesis lab
Once a reaction profile is understood the need for MS is reduced and can be just as well followed on a possibly more conveniently placed LC-UV system.
Alternatively MS only can be used to follow the progress of a reaction, i.e. disappearance of reagent and/or appearance of product without LC separation until such a point when a full LCMS is deemed necessary. Approaches for this include ambient MS interfaces such as DART, DESI and that offered by Advion which can be near instantaneous, and flow injection (loop) analysis. All of these techniques potentially suffer from ion suppression and no purity information is generated but they offer a fast readback that will frequently be fit for purpose. The siting of such systems relative to the synthesis labs is certainly an important factor in terms of the effective speed and convenience of the overall approach.
A further possibility is to use TLC/MS which gives both a separation and MS Information, though clearly the time to develop a TLC plate is significant. Better quality MS data might be expected thanks to the separation step. Systems from Advion and Camag are available if this is considered an attractive option.
High versus low pH
A catalyst-impregnated column will likely be less of an issue for low pH analysis compared to high pH, although clearly it is better to encourage (insist on) some form of work-up/extraction to remove the problem at source.
A Final Thought
One final strategy from a related application that came to mind while writing this article was an approach we developed at CombiPure to reduce the need for repeat QC analysis of final products ex-prep LC of compound libraries via optimisation of injection volume. We achieved this by means of pre-screening the samples in a UV plate reader. Aliquots of the final product solutions were diluted in acetonitrile:water [50:50] in 96 well plates and the UV spectrum was recorded in the reader. Via our LIMS the injection volume for final LCMS analysis was then fixed between 0.5 and 10ul using this data. This simple idea led to a 90% reduction in the amount of re-analysis necessary, a very significant time and cost saving in a high throughout environment. We similarly avoided over-injection of concentrated samples compared to using a default volume in the first instance (previously 3ul) and the associated problems.
Although all these strategies are technological in nature, it is worth noting here that it is always better when Analytical and Medicinal or Synthetic chemists work co-operatively to address issues of mutual interest and, as ever, the importance of communication, education and training must not be neglected. However, that is the start of another blog…
If any of the issues discussed in these articles are of interest to you, please feel free to contact me directly on email@example.com for further information and/or sign up at the bottom of this page to receive notifications of future articles and events.