Inconsistent Signal-to-Background for Li analysis in ICP-OES axial mode

Question

I have recently taken over the maintenance of a high-traffic (Varian) 720 axial ICP-OES. 
 Some of my users reported imprecise and inaccurate Li determination. In my troubleshooting, I have found that: 
 a) the intensities obtained from a known sample can vary extremely from sample to sample 
 b) the background can very quite a lot, further altering signal-to-background (SBR). 
 c) the replicates of a given measurement can vary quite a bit. 
 I have considered that this may be due to EIE effects, so I am suggesting to them to include an ionization buffer, which I believe (?) would help maintain consistent emission intensities from the Li lines. However, I am less certain about what to prescribe for the variable background, or what keywords to search for at this point with trouble shooting. 
 I am including some screenshots of the Li 460 nm peak from my investigation. I made one large sample of Li and simply split it into multiple tubes, Li 3_1, Li 3_2, etc. My calibration curve had great linear response and a great RSD...

Tom Herzog · Accepted Answer

You are on the right track with an ionization buffer. Cesium is a good option in your case. A lot of labs use Li for ionization buffer as it is cheaper, but then you can't analyze for Li. 
 Are you seeing the plasma become dimmer over the course of the run? It could be that the RF tube is getting old and the RF power is slowly dropping. This will both drop the background signal and increase Li intensity (Na and K as well). The higher the RF power, the more ionization of those Group I minerals and lower atomic emission intensity. Adding Cs will drive more of the ionized Li back to ground state and increase intensity. 
 If it is the RF getting weaker during the run, other wavelengths, especially those below 200nm, the background will drop but the intensity will drop as well because of the low power. Not sure if you analyze anything down there (As, Se, Sb, Tl, Al, Pb, Zn) but if you see their intensity dropping while your Li is increasing (without adding the ionization buffer) then your RF power is dropping due to a soft RF tube. 
 Plasma loading can also cause fluctuations in your signal. A partially salted nebulizer tip will aspirate too much mist into the plasma and cause variable signal. The TDS of a solution can affect the plasma as well. Organics can also cause variable signal. Do you use an internal standard? How does it recover when the background varies like you showed in the picture?

Inconsistent Signal-to-Background for Li analysis in ICP-OES axial mode

Top Replies