School of Engineering and Science - Jacobs University

20
15
BHVO-2 (n=5)
sample precision
run precision
method precision
%RSD
10
5
0
20
Sc Ti Co Ni Rb Sr Y Zr Nb Mo Cs Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta W Pb Th U
SGR-1b (n=2)
15
%RSD
10
5
20
0
Sc Ti Co Ni Rb Sr Y Zr Nb Mo Cs Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta W Pb Th U
FER-2 (n=3)
70.5% 34.9%
15
%RSD
10
5
0
30
25
Sc Ti Co Ni Rb Sr Y Zr Nb Mo Cs Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta W Pb Th U
JDo-1 (n=3)
36.3% 59.3%
%RSD
20
15
10
5
0
Sc Ti Co Ni Rb Sr Y Zr Nb Mo Cs Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta W Pb Th U
Figure 8. Comparison of average sample, run, and method precision for various CRMs, where n refers
to the number of HF-HClO 4 sample decompositions and ICPMS runs in which the CRM was measured
repeatedly. For most elements method precision is better than 5%, and is similar to, or even lower than,
the sample or run precision. This suggests that for these elements the standard deviation of the data is
primarily controlled by the inherent precision of the ICPMS instrument, and not by the decomposition
method. Poor method precision is also due to low abundances for some elements that approach the IQL,
e.g., Nb, Ta, and W in FeR-2 and JDo-1, as well as Co in JDo-1.
22