Wang 1994
Wang 1994
Wang 1994
inspection of its GC trace. For example, Figure 2 compares (n-czz + n-C24 + n-C26 + n-Cz8) (18),indicates the greater
the GC/FID traces (F3) for the samples S-6, S-2, and S-1. extent and degree of weathering of S-6 in comparison to
Figure 3 presents the GC/MS chromatograms of the m/z the weathered source oil. It is interesting to note from
85 fragment (one of the characteristic fragments for Figure 4 that due to the additional weathering, the relative
n-alkanes) of the saturated hydrocarbon fraction (Fl) of concentrations of n-alkanes before n-Cz6 in S-6 are lower
the weathered source oil, S-6 and S-4. The remaining than the corresponding n-alkanes in the weathered source
samples show almost identical GUMS chromatographic oil, and the n-alkanes after n-C26 show the reverse trend.
profiles as sample S-4. These GUMS chromatograms are The only exception before n-C26 is pristane and phytane
not particularly useful for determining the source of the with the relative concentrations in S-6 being significantly
spill, but they give information on the degree of weathering higher than in the weathered source oil, which is an
or freshness of the sample, which is indicated by the indication that the isoprenoids pristane and phytane were
distribution of n-alkanes and by the profile of the hump. degraded at a slower rate in comparison to n-alkanes. A
Figures 2 and 3 clearly demonstrated the effects of preferential loss of n-alkanes often indicates the occurrence
weathering and degradation on the chemical compositions of biodegradation (19).
of the oils. The GC/MS chromatogram of the saturated For S-4 and the remaining samples, only unresolved
hydrocarbons in the weathered source oil (Figure 3A) shows envelopes or humps were observed in their GC traces. By
a mixture dominated by a homologous series of n-alkanes visual comparison, these broad humps are very similar in
ranging in carbon number from n-Cg to n-C4o with the the distribution patterns, indicating that they were highly
maxima around n-Czo-n-Cz~. Isoprenoid hydrocarbons weathered and possibly came from the same source.
which are commonly considered in weathering studies are Obviously, such GC traces cannot provide much informa-
also present, and pristane and phythane are the most tion about compositional changes nor be very useful for
abundant. It is suprising to note that the existence of the purpose of matching the samples to the known source
what is apparently original Arrow oil after 22 years in S-6 oils.
sampled from pools on a beach has characteristics almost Quantitation of Target PAHs and Their Alkylated
identical to the same oil stored in a sealed container (i.e., Homologues. PAH compounds are relatively stable and
the weathered source oil). In sharp contrast, the n-alkanes diagnostic constituents of petroleum. Some studies have
including the five less degradable isoprenoids (2,6,10- been done using the distribution of the alkylated PAH
trimethyldodecane, 2,6,10-trimethyltridecane, norpris- homologues as environmental fate indicators and source
tane, pristane, and phytane) in the sample S-4 and other specific markers of oil in sediments (17,20,21)and tissue
samples were nearly completely lost, indicating that these samples (22, 23).
samples have undergone extreme alternation. Another Table 4 summarizes the quantitation results of alkylated
pronounced feature for the S-4 and other samples is the PAH homologues in the Arrow oil samples. Figure 5
big humps representing the GC unresolved complex depicts graphically the PAH distributions of the weathered
mixture of hydrocarbons in the chromatograms, which is source oil, S-6, S-A, and S-9. In order to more clearly see
characteristic of heavily weathered and biodegraded oil the changes in the distribution patterns within the different
(17). PAH families, the different scales for Y-axis were applied
Table 3 presents the quantitation results for n-alkanes to Figure 5A (0-3200), 5B (0-800), and 5C and 5D (0-400
in the weathered source oil and S-6. Figure 4 compares yg/g TSEM). The aromatic fraction of the weathered
the n-alkane distribution patterns from these two samples, source oil contains primarily alkylated naphthalene and
based on the amount of the n-alkane c26 (cZ6is selected phenanthrene homologues with CZ-and Cs-naphthalenes
as it is at the point in the series where the relatiev changes being dominated (3073 and 3013 pg/g, respectively).
go from loss to gain). The concentration reduction of the Alkylated dibenzothiophenes are also present, but in lower
low boiling components and the buildup of the high boiling concentrations than the phenanthrene series. The loss of
components relative to the smaller n-alkanes indicate that PAH compounds and their alkyl homologues in the Arrow
the source oil and S-6 have been significantly weathered, oil samples is very apparent. The factors affecting the
even though the n-alkanes in these two samples were not weathering process of the spilled oil in the environment
completely lost. The lower value of the weathering index during the past 22 years are very complex (24). The
(WI),defined as the ratio of (n-Cs+ n-Clo + n-Clz + n-C14)/ combined effects of weathering (such as evaporation,
1 .sed
1.4e4-
12e43
10-4.-
EOOO-,
4
6000-
4000-
2ooo+-
0 20 40 60
1 , ’
0 1 0 20 30 40 50 60 7 0
2.2’q C
2.0’43
1.2-4 IS
8000-1
60004
4000j
0 20 40 60
Flgure 2. GC/FID traces for the Arrow oil samples S-6 (A), S-2 (B), and S-1 (C). IS means internal standard.
ic000 -
1
15000.
, lG000!
~ 5000; !I
pbundance Ion 85.00
30000 ! B
25000 -I
20000 -! I
, 15000,
:: I
0 I
I
10000
,
5000
0
Tire--> 5.00 10.00 15.00 20.00 25.00 30.00 35'00 40.00 45.00 50.00 55.00
1 250COj
1
10000 -
, 5coc-
Flgure 3. GC/MS chromatograms of saturated hydrocarbons (mlz 85) for the weathered source oil (A), S-6 (B), and S-4 (C). Numbers represent
carbon numbers of +alkanes. i-C15, 1416, 1418, Pr, and Ph represent the five most abundant isoprenoids (2,6,10-trimethyidodecane, 2,6,10-
trimethyitridecane, norpristane, pristane, and phytane, respectively).
dissolution, photooxidation, and biodegradation) of the PAHs. Major compositional changes of PAHs observed
spilled oil may be responsible for the progressive loss of are as follows: (1)pronounced decrease in naphthalene
".Ab".,
Flgure 4. Comparison of t b relstlve Raikane concentrations and the ~ h n distribution
a patterns for Uta weamered m c e oil and the sample
S-6. The arrows indicate that before the reletbe concentrations of pristane and phylane in the sample S-6 are h i g k than in the weathered
source oil.
PAH Homologues
Flgure 5. PAH flngerprints of the weathered source oil (A), S-6 (B), S-A (C), and S-9 (D) illustratingthe effects of weathering on PAH compositions.
N, P, D,F, and C represent naphthalene, phenanthrene, dibenzothiophene, fluorene, and chrysene, respectively; 0-4 represent carbon numbers
of alkyl groups In alkylated PAH homologues.
09-
08-
-
-i-
ChryrIPhens
ChryslDibenr
abundant and even undetectable. More importantly, the
ratios of five pairs of target hopane compounds [Ts/Tm,
07- c 2 3 / c 2 4 , c 2 9 / c 3 0 , c 3 2 20S/(20S+20R), and C 3 3 2 0 8
06- (2OS+2OR)] are almost the same (Table 6) for most
samples. These facts strongly suggested that they may
come from the same source.
The ratio of Ts to Tm has been used as a geochemical
parameter to characterize both source input and maturity
of a crude (25) and as a indicator for oil matching (26).
0% ~ _ _ - However,in many cases, these two peaks are not abundant
D vi
4
enough to give a precise and accurate ratio of Ts/Tm. We
: found in this work that the ratios of two pairs of the most
Figure 6. Plot of the relative ratios of the total alkylated chrysenes
abundant hopanes (C23/C24 and c 2 9 / c 3 0 ) , especially the
to phenanthrenesand dibenzothiophenesversus the Arrowoil samples.
ratio of the hopanes c 2 9 / c 3 0 , can be used as very good
to draw any positive conclusion on whether these 22-year- indicators for oil matching and oil source tracking. Table
old samples came from the same source. Fortunately, the 6 shows that the values of Ts/Tm range from 0.26 to 0.29,
fingerprinting of triterpanes provides us with a powerful the values of c 2 3 / c 2 4 and c 2 9 / c 3 0 range from 1.98 to 2.10
tool for tracking the source of the long-term weathered and from 0.84 to 0.87 for samples from S-A to S-9,
oil. In this work, two criteria were used for matching the respectively. But for samples S-1and S-10,the values of
long-term weathered oil with the source oil: (1)whether Te/Tm, c 2 3 / c 2 4 and c 2 9 / c 3 0 are 0.31 and 0.36, 0.62 and
the distribution profile and pattern of m/z 191 and 217 1.00, and 0.87 and 0.89, respectively. Among them, only
fragments are the same and (2) whether the computed the c 2 9 / c 3 0 ratios appear to be consistent and to be the
ratios of some target pairs of hopane and sterane com- best indicator to track the oil source for these two most
pounds are near identical. highly weathered samples. Beside that, the relatively lower
The GUMS measurements demonstrate that the pro- abundant but well-resolved C 3 2 (peaks 25 and 26) and C 3 3
files and patterns of the m/z 191 and 217 ion chromato- (peaks 27 and 28) 20S/20R hopanes also show very good
grams of the Arrow oil samples are nearly identical, except consistency in thevalues of 2OS/(2OS+2OR) ratios, ranging
that the samples S-1 and S-10 showed the similar from 0.57 to 0.61 for c 3 2 bishomohopanes and from 0.60
distribution patterns but with some peaks being less to 0.64 for C 3 3 trishomohopanes, respectively.
J
I
I
~ 3500
!I
6
19
2000 4
1 5
4
Isi
22
1500
2
-1
~
1000
1 0
IS
3000 1 20 I
1 19
2500 1
0 i " " I " " l " " / " " l " " I "
ime-->25.00 30.00 35.00 40.00 45.00 50.00
~
Flgure 7. Distribution of triterpanes (mlz 191) for the weathered source oil (A) and S-4 (B). The right inset in panel A is the chromatogram of
m/z 177 in 38-42 min. See Table 6 for component identification. The peaks 14 and 17 are identified as C27 254risnorhopane and CZ825-
blsnorhopane, which show significantly higher abundance of the mlz 177 fragments. The general structure for hopanes is shown in the lefi inset
of panel A.
Under most natural weathering conditions, which usu- terpanes and of higher carbon number vs lower carbon
ally has evaporation as the main process, the Ts/Tm ratio number pentacyclic terpanes. This feature is very valuable
should suffer little change, if any, and remain characteristic in studies on the behavior and fate of spilled oil in the
of the source crude. The same is true for C23/C24. But, environment and can be used to unambiguously distin-
under very long and extreme weathering conditions, the guish one highly weathered oil from another. For S-1and
ratios of Ts/Tm and c 2 3 / c 2 4 may suffer change. I t has S-10,the fact that they have the much lower C23/C24 values,
also been found that Tm has a relatively faster rate of the lower Ts/Tm values, and almost the same c 2 9 / c 3 0
depletion than Ts in the process of crude maturation (26). values, as well as the same C 3 2 2OS/(2OS+2OR) and C 3 3
In contrast, the ratios of C29/C30, c 3 2 2OS/(2OS+2OR),and 20S/ (20S+20R) values, strongly suggests that they may
C 3 3 20S/(20S+20R) appear to be unaltered, which may come from the same source, but were very highly weath-
reflect differential weathering of pentacyclic vs tricyclic ered.
Table 6. Analysis Results of Selected Hopanes and Steranes in Arrow Oil Samples
hopanes steranes
CW@- CWB-
c23 triterpanes Ca triterpanes ( p g / g of (pglg of C23/ Ts/ Cm/ c32 s/ C33 SI czsa@@/ czs a@@/
sample ( p g l g of TSEM) ( p g / g of TSEM) TSEM) TSEM) C24 TmO Cm (S+R) (S+R) (a@@+ a 4 (a@@ +
weathered 104 50 91 106 2.07 0.26 0.86 0.60 0.62 0.55 0.52
sourceoil
S-A 146 71 125 145 2.05 0.27 0.86 0.58 0.63 0.58 0.52
s-2 122 59 112 134 2.05 0.28 0.84 0.60 0.63 0.56 0.53
5-3 203 99 157 183 2.06 0.26 0.86 0.58 0.62 0.57 0.51
s-4 147 67 117 136 2.07 0.27 0.86 0.58 0.61 0.56 0.52
5-5 121 57 115 132 2.10 0.27 0.87 0.57 0.62 0.56 0.52
S-6 115 57 92 107 2.06 0.26 0.86 0.59 0.62 0.55 0.51
s-7 166 79 125 143 2.09 0.27 0.87 0.57 0.63 0.56 0.53
S-8 153 74 118 137 2.07 0.28 0.86 0.57 0.62 0.58 0.51
s-9 77 39 104 120 1.98 0.29 0.86 0.61 0.64 0.56 0.53
s-1 5 8 32 37 0.62 0.31 0.87 0.60 0.61 0.60 0.54
s-10 40 40 53 60 1.00 0.36 0.89 0.57 0.60 0.59 0.52
Ts: 18a(H),21fl(H)-22,29,30-trisnorhopane;
Tm: 17a(H),21@(H)-22,29,30-trisnorhopane.
I 0 1 , I , I , , , , , , , , , , , , , , , ,
bime--> 32;OO 34100 36100 38:OO 40.'00 ' 42;OO 44100 ' ' 46,'OO ' 48.00 '
- 0 I " ~' I ' " I " ' I " " I " " I " ' I "
ime--> 32.00 34.00 36.00 38.00 40.00 42.00 44.00 46.00 48.00
Flgure 8. Distribution of steranes (mlr 217 and 218) for the weathered source 011 (A) and S-4 (B). See Table 6 for component identification.
The dashed lines show a:,b,b-steranes which glve higher abundances of the mlr 218 fragment than the mlz 217 fragment. The general structure
for steranes Is shown in the right Inset of panel A.
It is evident that the most important advantage using relative intensities of m/z 149-151 fragments were used
c29/c30 as the source indicator is that C 2 9 and c 3 0 are the to distinguish between 5-a: and 5-p stereo isomers. It has
most degradation-resistant compounds compared to C23/ been recognized that the only significant difference
C24 and Ts/Tm. Secondly, they are often the most between 5-a: and 5-0 isomers is that the m/z 149 fragment
abundant among hopanes from C 1 9 to C36; therefore, the is more abundant than the m/z 151 moiety of the 5-a isomer
c 2 9 / c 3 0 ratio can be most accurately computed. So far, (23,27). Furthermore, the retention time of the 5-@isomer
hundreds of oil samples have been analyzed in our is shorter than that of the 5-a: isomer. In addition, it was
laboratory. The combination of these ratios, especially found from studying the paired standard compounds of
C z g / C 3 0 a s well C 3 2 2OS/(2OS+2OR) and C 3 3 20S/(20S+20R), C27, (328, and C 2 9 a:,a:,a:- and a,P,P-steranes that the P,a,a:-
has been satisfactorily applied to match and identify oils. and a,a,a-steranes have a base peak at m/z 217, while
For sterane identification, in addition to the base peak a,P,P-steranes have a base peak at 218. This feature,
chromatograms of steranes at 217 and 218 (Figure 8), the combined with the measurement of the ratio of the m/z