Muhamad Sufiyan, Azah
(2009)
Discrimination of diesel by gas chromatography.
Other.
Universiti Sains Malaysia.
(Submitted)
Abstract
Diesel samples from 27 service stations in Kota Bharu were collected over a period of a week
were subject to GC-MS analysis. Using extracted ion chromatograms, peak area of 41
selected peaks at various mass per charge were integrated. These peak areas were then selfnormalised
to ion 220b prior to Principal Component Analysis (PCA).
The results show that principal component analysis is able to group the samples into a few
clusters. Using the distance between replicates, however, some samples are tightly group
indicating that these samples are relatively similar based on the data treatment and statistical
analysis used in this study.
A blind test matched the target sample indicating that the data treatment is reliable. The
reliability of this approach is further supported by the close proximity between duplicates.
Four samples collected from a selected service station over a period of four weeks were
spread out by PCA indicating that these diesel profiles change over time. This is true if the
service station received new batches of diesel after each collection of the samples.
Actions (login required)
 |
View Item |
