Hi All,
We are currently trying to validate our QQQ, for a multi element fusion digest method, against a 7900 ICPMS. Both instruments use the same sample introduction system, and the same standards, however, I cannot get the QQQ to behave in the same way as the 7900. The matrix is ~3% nitric acid with lithium meta/tetra borate flux.
I've attached a table below showing the sort of thing that I'm seeing from the 8900, looking here at lanthanum mean average percentage difference, using a large range of geological standards from 20ppm to 2.2%. The same solutions and standards were run on both the 7900 and the QQQ, three times and then compared. For lanthanum listed here the 7900 is running in He mode and the QQQ in N2O (He mode was much worse)
What I don't understand is why I get a taper off over the concentration range. This corresponds to either an increased concentration reported by the QQQ at higher concentrations, or possible under reporting at the lower concentration ranges. If I compare the same samples run by OES, the 7900 and OES compare well, while the same curvature exists between the QQQ and the OES, pointing to this as an issue with the QQQ.
Originally, we had Ce and La set to run by He mode on the QQQ, but after some experimenting, we found that N2O performed better (He was worse, see below), but still, it is not a flat concentration curve comparison. Extending the element suite showed that Ce, La, Gd, Nd and Pr all show this same characteristic.
In some of the earlier work, we often found step changes in the data, around the 0.06ppm mark (as per the graph below, Ce (He)). This was solved mostly by moving to N2O reaction gas, but this did point me to think it was a PA factor issue that I might be dealing with. However, since mass hunter now automatically corrects for PA factor in the batch using calibration standards, I'm not sure where to go from here.
Has anyone else seen this sort of behaviour and found a solution to it?
