MS/MS sensitivity loss

[tayzyboy]

Hi,
We have a Waters Quattro premier + UPLC.
Mobile phase is 75% H2O, 25% MeCN +0.1% Formic acid. 0.3ml/min
Column is an Aquity C18
Runtime is 16minutes - I know this is long, but we are separating isomers.

The problem is that over a series of injections of compound diluted in mobile phase, we get a gradual loss in sensitivity. If 100 samples are run, the last one will be roughly 10% of the first.
When the machine is left to run mobile phase over night without injecting, the sensitivity returns on re-injection next morning, so it is not a stability problem.
Reading data from the UV absorbance detector has good consistancy, showing it cant be the injection system.
We added an internal standard (IS sensitivity also drops), however the massive loss in sensitivity means our method would not be suitable for analysis of biological extracts, we need to keep the sensitivity up!.
We have also tried diverting the first 5minutes of the run to waste, incase there were any impurities fouling the MS, but that hasnt helped either.

Any ideas would be appreciated.
Thanks
James

[tayzyboy]

oh and...

MS settings are roughly:
3.5 cap V
35V cone
150 source temp
300 desolv temp

20 collision

Mass of compound is roughly 500m/z with transition to aprox 150m/z

[bert]

Hello james,

would be interesting to know what your sample composition is?

regards Bert

[trackster]

Hi James,
I am not familiar with this instrument in particular, but tend to agree with you that it sounds like some sort of instrument fouling by the sample injections. Can you limit what is allowed to go to the MS further than just the early portion? i.e. only allow the components of interest and a slight window before and after to allow for some RT drift?

Could also try putting in some good sized injections of pure MeCN or MeCN with 0.1% formic after every so many samples to see if that helps.

Only other thought is that maybe conditions aren't optimal for the mobile phase and the source is getting flooded a bit. As I said, I am not familar with this instrument or parameters and you likely have run other similar methods on it, so I doubt this is the case. Additionally, you noted that running mobile phase through it overnight seemed to restore sensitivity, which would not support this idea (assuming that flow is at same rate). If flow was much lower, could have allowed source to dry out over time again.

Just a couple thoughts. Not sure that they are any help, but all I have not knowing this instrument better and other details.

Good luck!


Randy

[tayzyboy]

cheers for the ideas, im still working on it.

The sample is pure compound (well 99%+) dissolved in DMSO then diluted in mobile phase. DMSO concentration in an injection is 0.01%.