The Xevo TQ-XS system's connected to Waters UPLC with Methanol based ion pairing (HFIP/DIEA) gradient at low flow rate around 200 ul/min. The analyte is an oligonucleotide under denaturing condition (high LC column oven temp). The gradient time (minus washing/re-equilibration) is around 6 minutes. Under MRM mode (or even for full scan mode), there's a nice peak but as the baseline falls down from the apex, it doesn't fall all the way down but stays elevated for a while (looks like below:)

Image

(Sorry for the ugly drawing but I cannot post any confidential info)

I've seen other abnormal peak shapes but have never seen anything like this.

I tried changing gradient time (shorter and longer), flow rate but to no avail. Shorter gradient time actually seemed to make the 'plateau' part go up a bit.

Any suggestion would be appreciated!