Interference reduction collision/reaction interface

The other thing to be wary of with semiquantitative analysis is the spectral complexity of unknown samples. If you have a spectrally rich sample and are not making any compensations for spectral overlaps close to the analyte peaks, it could possibly give you a false-positive for that element. Therefore, you have to be very cautious when reporting semiquantitative results on completely unknown samples. They should be characterized first, especially with respect to the types of spectral interferences generated by the plasma gas, the matrix, and the solvents/acids/chemicals used for sample preparation. Collision/reaction cells/interfaces can help in the reduction of some of these interferences, but extreme care should be taken, as these devices are known to have no effect on some polyatomic interferences, and in some cases can increase the spectral complexity by generating other interfering complexes. 

---