Summary and Conclusion 
33 
FID 
MS 
IRMS 
RWS 
BSH 
IAEA 
NFI 
RWS Height BSH Area BSH Height 
IAEA NFI 
RSD CD RSD CD RSD CD RSD CD 
RSD CD RSD CD RSD CD 
1 %0 1 %0 
Set_l 1 on 2 71% 71% 47% 60% 62% 69% 58% 75% 
1 on 3 57% 71% 60% 80% 69% 69% 83% 83% 
2 on 3 64% 71% 53% 60% 69% 69% 50% 67% 
Set_2 4 on 5 45% 45% 35% 48% 53% 59% 29% 35% 
4 on 6 20% 15% 43% 52% 24% 47% 24% 29% 
5 on 6 25% 20% 30% 39% 29% 29% 12% 18% 
Set_3 7 on 8 65% 100% 94% 94% 75% 60% 100% 100% 
7 on 9 53% 65% 67% 89% 58% 87% 64% 71% 
8 on 9 59% 82% 83% 94% 75% 67% 71% 71% 
Dup 1 on 10 100% 100% 100% 100% 100% 100% 100% 100% 
64% 71% 45% 
86% 86% 73% 
71% 71% 55% 
64% 
82% 
64% 
50% 50% 
42% 67% 
50% 67% 
45% 25% 
30% 25% 
35% 30% 
22% 
17% 
9% 
30% 28% 44% 
22% 11% 28% 
13% 17% 28% 
100% 100% 100% 100% 36% 71% 
59% 71% 64% 79% 29% 50% 
65% 82% 57% 79% 64% 79% 
86% 71% 100% 100% 91% 82% 
100% 83% 
100% 100% 
100% 100% 
8% 31% 
0% 
0% 
92% 92% 
77% 92% 
85% 92% 
83% 100% 
CD = Critical Difference (14%) 
RSD = Relativ (Mean) Standard Deviation (5%) 
Tab. 6: Total conclusion matrix of all analytical approaches (GC-FID. GC-MS, GC-IRMS). 
The statistical method suggested by René der Bruyn (NFI), which compares 
standard deviation vs. weighted average, might work best for the assessment 
of data derived from a single laboratory. The results of the presented study 
show differences in Inter-laboratory comparison which are most likely related 
to very Individual properties within different GC-FID systems as the patterns 
of data sets from different laboratories are comparable, but having an offset 
on both axes. To make results more comparable between different laborato 
ries, an additional normalisation on a common standard (paraffin-wax) might 
be beneficial.