internal standard troubles

We've had halted production on our ICP-OES and have finally received a whole back up consumables kit and have replaced those items we evaluated during our trouble shooting process (new nebulizer, autosampler probe, water chiller etc.) Before going offline we had decided to change to a smaller capillary size on the I.S. p-pump tubing to prevent the need for excessively remaking solution/accumulating waste. As it stands things are just about up to speed, however it seems that the first couple calibration points have a stable I.S. response after which the response drops 20% +/-. I'm wondering if maybe the uptake time needs to be adjusted? The new autosampler line is also longer then the one we put to rest so maybe that can be taken into account. 

I just feel some where in the sample introduction the solutions are not being mixed well or that the timing has been thrown off with the adjustment in tube length, capillary size, etc. 

Any advice regarding this will be helpful.

Parents Reply
  • The smallest internal standard tubing that I have used is Orange/Green 0.38mm, I have never used the orange-blue tubing.   What is the matrix of your calibration standards? If they have high concentrations of metals then the internal standard will decrease due to suppression, as the concentrations get higher. 

Children
  • We run a 5 point calibration range from 0.1ppm - 2.0ppm for for (As, B, Co, Cr, Pb, Sn, Cd, Be, and Ni) in a 5%HNO3 matrix including a blank. I had completed a calibration after extending the uptake delay another 5 seconds with notable results. Looks like were closer to a 10% variance which is closer to our acceptable limit. I mean to adjust the stabilization time another few seconds and see what happens.

    [Update] still conflicting with the blank IS response and the rest of the calibration run. All cal points are averaging about 4k counts lesser than the cal blank but remain stable.

    [Update} made fresh calibration and QC standards, reset the original stabilization time, and IS response is stable and within limits. I had started off the calibration with an old (concentrated) matrix blank and re-pouring fresh standards seem to do the trick.

Was this helpful?