Analysis of Hydrocarbon-bearing Fluid Inclusions (HCFI) using timeresolved fluorescence spectroscopy
Milosz A. Przyjalgowski,
*a,b
Alan G. Ryder, Martin Feely, and Thomas J. Glynn
b,c
d
a
Department of Physics, National University of Ireland, Galway;
National Center for Biomedical Engineering Science, National University of Ireland, Galway;
Department of Chemistry, National University of Ireland, Galway;
Department of Earth and Ocean Sciences, National University of Ireland, Galway;
a
b
c
d
ABSTRACT
Hydrocarbon-bearing fluid inclusions (HCFI) are microscopic cavities within rocks that are filled with petroleum oil, the
composition of which may not have changed since the trapping event. Thus, the composition of that entrapped oil can
provide information about the formation and evolution of the oil reservoir. This type of information is important to the
petroleum production and exploration industries. Crude oil fluorescence originates from the presence of cyclic aromatic
compounds and the nature of the emission is governed by the chemical composition of the oil. Fluorescence based
methods are widely used for analysis of crude oil because they offer robust, non-contact and non-destructive
measurement options. The goal of our group is the development of a non-destructive analytical method for HCFI using
time-resolved fluorescence methods.
In broad terms, crude oil fluorescence behavior is governed by the concentration of quenching species and the
distribution of fluorophores. For the intensity averaged fluorescence lifetime ( τ ), the best correlations have been found
between polar or alkane concentrations, but these are not suitable for robust, quantitative analysis. We have recently
started to investigate another approach for characterizing oils by looking at Time-resolved Emission Spectra (TRES).
TRES are constructed from intensities sampled at discrete times during the fluorescence decay of the sample.
In this study, TRES, from a series of 10 crude oils from the Middle East, have been measured at discrete time gates (0.5
ns, 1 ns, 2 ns, 4 ns) over the 450-700 nm wavelength range. The spectral changes in TRES, such as time gate dependent
Stokes’ shift and spectral broadening, are analyzed in the context of energy transfer rates. In this work, the efficacy of
using TRES for fingerprinting individual oils and HCFI is also demonstrated.
Keywords: crude oil, petroleum, hydrocarbons, fluorescence, Time-resolved Emission Spectra (TRES), energy transfer,
fluorescence lifetime, hydrocarbon-bearing fluid inclusions (HCFI).
1. INTRODUCTION
Hydrocarbon-bearing fluid inclusions (HCFI) occur in rock and vein forming minerals, e.g. quartz and calcite, and
contain micron scale samples of trapped hydrocarbon oil. In addition to hydrocarbon fluids, at room temperature, HCFI
can also contain traces of other fluids such as H O, gasses such as CO and CH , and solids that were either entrapped
along with/or precipitated from the liquid phase at later times. The accurate determination of the composition of the
trapped oil in single fluid inclusions can yield (together with micro-thermometric studies) fundamental information about
the history of oil formation and migration in sedimentary basins, which is important to both the petroleum production
and exploration industries and to the geological community.
Destructive analysis of HCFI can be done either as bulk ensemble analysis or as a single inclusion analysis. The bulk
fluid inclusion analysis is typically done by crushing rock samples and extracting the fluids trapped in inclusions for
chromatographic analysis. The oil composition data obtained this way may suffer from variety of problems including
mixing of fluids from multiple nonhomogenous HCFI populations and contamination from material within the rock
sample itself or lab contamination. Analysis of single inclusions is, in principle, much better. It requires opening of HCFI
(laser ablation or high energy beam ions), but the very low quantities of oil typically contained in HCFI (φ 10µm ~10 g)
restricts analysis with gas chromatography and/or mass spectroscopy to rather large inclusions or groups of inclusions.
Optical micro-spectroscopy, on the other hand, offers good diagnostic potential for non-destructive HCFI analysis. The
most common spectroscopic methods are based on infrared absorption , Raman, and fluorescence spectroscopy.
1
2
2
4
2
3-5
6,7
-9
8,9
10,11
*
Author to whom the correspondence should be sent: milosz@nuigalway.ie
12
3,13
UV fluorescence is commonly used for identification of HCFI
and the high sensitivity of fluorescence and wide
range of available techniques makes it a potentially viable tool for HCFI analysis.
Fluorescence of petroleum products, which originates from the presence of polycyclic aromatic fraction, is directly
governed by chemical composition and physical properties of the oil. In general, heavy oils (low API gravity) tend to
have relatively broad, more shifted to the red and less intense fluorescence bands, lower quantum yields, and shorter
fluorescence lifetimes as compared to light oils (high API gravity). These effects are produced by higher concentrations
of fluorophores and quenchers in heavy oil, which yields higher rates of energy transfer and quenching.
The use of fluorescence color is widely used as a qualitative guide for assessing the maturity of crude oil in bulk and in
but this technique is intrinsically prone to error. In principle, all techniques based on measurement of
HCFI,
fluorescence intensity and spectral distribution can be adversely affected by instrumental factors such as excitation
source intensity variation, detector spectral sensitivity variation, and also by the physical properties of sample such as
opacity, geometry, turbidity, scattering properties, and photobleaching.
Alternatively, time-resolved based fluorescence techniques are not only largely free of these artifacts, but also contain
information that is lost in time-averaging processes inherent to steady-state methods. These techniques have also been
The early studies were mainly
used to characterize various petroleum products down through the last 30 years.
driven by a need to develop a tool for monitoring the environmental pollution, with no attempts made to correlate
The most commonly used descriptor in time-resolved studies of
experimental data with chemical composition.
petroleum oils is fluorescence lifetime, which is also governed by the complex interplay between the energy transfer and
quenching, with heavy oils having typically shorter lifetimes then light oils.
The fluorescence lifetime is
emission/excitation wavelength dependent since each emission/excitation wavelength represents different populations of
fluorophores.
In our laboratory, we have been investigating the fluorescence behavior of crude oils with the idea of developing a
robust, quantitative method for oil characterization using mainly the intensity averaged fluorescence lifetime ( τ ).
Recently, we have measured fluorescence lifetimes from a series of HCFI and correlated the data with those obtained
from 23 crude oils. The best correlations have been found between τ and polar or alkane concentrations. These results,
although not suitable for highly accurate quantitative analysis, allow us to classify the entrapped oils according to polar
and alkane concentrations. In this previous study the oils were sampled from diverse sources and locations, and their
extensive geological range contributes to the scatter in calibration models for oil composition.
1,3,14
15
16
†
17
18
19
16, 20-23
24
3,5,13
25,26
27
15
19,23,28
28-30
19,20
20-23,31
32
t1
t2
t3
t1
yt
is
n
et
in
tyi
s
n
et
n
i
t2
t3
wavelength[nm]
time[ns]
Schematic diagram of Time-resolved Emission Spectra (TRES). TRES are obtained during the time windows denoted by t1,
t2, t3. For graphic simplicity only a single decay curve is shown.
Figure 1:
In this work, we investigate another approach for characterizing oils by conducting a detailed time-resolved fluorescence
study on a set of 10 well-characterized crude petroleum oils from a single geographical locality (Middle East) using
The API gravity is a measure of density and is related to the density by the formula:
API gravity=((141.5/specific gravity at 15.6°C)-131.5); i.e. H2O=10˚ API.
†
Time-resolved Emission Spectra (TRES). Conceptually, TRES are fluorescence emission spectra constructed from
intensities sampled at discrete times during the fluorescence decay of the sample (Figure 1). TRES have been previously
demonstrated to have potential as a tool for discriminating different oils types. However, neither any fundamental
studies into crude oil photophysics nor any correlations of spectral parameters with chemical composition were reported
in these studies.
Here, the correlations between spectral parameters such as time-gate dependent batochromic shift, spectral broadening
and chemical composition are examined in the context of fluorescence energy transfer rates. In addition, we also
demonstrate the potential for using TRES for the fingerprinting of the individual HCFI and matching the data with bulk
oils.
33,34
2. MATERIALS AND METHODS
2.1. Apparatus and procedures
Fluorescence data from bulk oils and the inclusion samples were recorded using the Diode Laser Fluorescence Lifetime
Microscope (DLFLM) developed in-house and described in detail elsewhere. In brief, this Time Correlated Single
Photon Counting (TCSPC) system is built around an Olympus BX-60 microscope and a 405 nm laser diode. The
instrument is capable of measuring fluorescence lifetimes (0.1 ns+), steady-state and Time-resolved Emission Spectra
(TRES) over the wavelength range 450-800 nm.
All bulk oil studies were carried out on neat non-degassed crude oil, held in 1 mm path-length, quartz cuvettes at room
temperature. The cuvettes were solvent and acid washed between the samples to prevent any cross contamination.
The fluorescence emission spectra were corrected using wavelength–dependent output of a tungsten filament calibrated
relative to black body radiation.
Fluid inclusions measurements were carried out on non-epoxied wafer following the procedure described by Ryder and
Przyjalgowski
.
32,35
36
et al.
32
2.2. Materials
The crude oil samples were provided by the Petroleum Geochemistry Division of Fugro Robertson (FR), Llandudno,
North Wales. The crude oil samples originate from various different locations across the Middle East and represent two
different oil reservoir intervals. The crude oils were analyzed in FR laboratory by column chromatography after removal
of the volatile hydrocarbon fraction by topping. The concentration of alkane, polar and aromatic fractions obtained from
chromatographic fractionation do not sum to 100% because of the column losses. Therefore the compositional data were
normalized to obtain corrected concentrations. The most salient data about the oils is given below in Table 1.
Table 1:
Oil
ME-1
ME-2
ME-3
ME-4
ME-5
ME-6
ME-7
ME-8
ME-9
ME-10
Chemical composition, physical properties and source rock information for the crude oils used in this study.
Reservoir
interval
Cretaceous
Cretaceous
Cretaceous
Cretaceous
Cretaceous
Jurassic
Jurassic
Jurassic
Jurassic
Cretaceous
API
[deg]
32.7
28.9
30.8
39.2
44.1
39.6
42.1
44.2
47.8
24.7
Asphaltene
[%]
3.4
4.2
3.0
0.9
0.1
0.1
0.0
0.04
0.0
8.9
Corrected
alkane [%]
61.6
61.7
62.1
71.1
80.4
76.3
80.6
77.0
84.8
51.2
Corrected
aromatic [%]
29.8
30.3
30.3
22.1
15.9
19.7
16.8
21.2
12.7
41.4
Corrected
polar [%]
8.6
8.0
7.6
6.7
3.7
4.0
2.7
1.9
2.5
7.4
Sulfur
[%]
1.63
1.43
1.67
0.92
0.60
0.58
0.83
1.13
1.80
2.67
The fluorite hosted hydrocarbon-bearing fluid inclusion sample F3a originated from Cave-in-Rock mining district,
Illinois, USA. 32
3. RESULTS AND DISCUSSION
3.1.
Bulk oils studies
Figure 2 shows Time-resolved Emission Spectra (TRES) of ten Middle Eastern (ME) crude oils measured for 0.5 ns, 1
ns, 2 ns and 4 ns time gates over 450–700 nm wavelength range.
]u
.a
[
tyi
sn
ten
i
de
zli
a
rm
on
A
1.0
0.8
.]u
.a
[y
ti
sn
ten
i
de
zli
a
rm
on
B
1.0
0.8
0.6
0.6
ME-10_0.5ns
ME-10_1ns
ME-10_2ns
ME-10_4ns
24.7=API
0.4
0.2
ME-2_0.5ns
ME-2_1ns
ME-2_2ns
ME-2_4ns
28.9=API
0.4
0.2
0.0
0.0
450
]u
.a
y[it
nse
tn
i
de
zli
a
m
ro
n
500
550
600
wavelength[nm]
650
700
.]u
.a
[y
ti
sn
et
in
de
zli
a
rm
on
C
1.0
450
500
550
600
wavelength[nm]
650
700
D
1.0
0.8
0.8
0.6
0.6
ME-3_0.5ns
ME-3_1ns
ME-3_2ns
ME-3_4ns
30.8=API
0.4
0.2
ME-1_0.5ns
ME-1_1ns
ME-1_2ns
ME-1_4ns
32.7=API
0.4
0.2
0.0
0.0
450
500
550
600
wavelength[nm]
650
700
E
1.0
]u
.a
[y
its
ent
ni
ediz
la
m
ro
n
0.6
500
550
600
wavelength[nm]
700
F
ME-6_0.5ns
ME-6_1ns
ME-6_2ns
ME-6_4ns
API=39.6
0.8
0.6
0.4
650
1.0
].
u.
a[
yit
nse
tn
i
edz
lia
m
ro
n
ME-4_0.5ns
ME-4_1ns
ME-4_2ns
ME-4_4ns
API=39.2
0.8
450
0.4
0.2
0.2
0.0
0.0
450
500
550
600
wavelength[nm]
650
700
G
1.0
ME-7_0.5ns
ME-7_1ns
ME-7_2ns
ME-7_4ns
API=42.1
.]u
.a
[y
its
ent
in
de
zil
a
m
ro
n
0.8
0.6
450
500
550
600
wavelength[nm]
ME-5_0.5ns
ME-5_1ns
ME-5_2ns
ME-5_4ns
API=44.1
1.0
.]u
.a
[y
its
ent
in
de
zil
a
m
ro
n
0.8
0.6
0.4
650
700
H
0.4
0.2
0.2
0.0
0.0
450
500
550
600
650
wavelength[nm]
ME-8_0.5ns
ME-8_1ns
ME-8_2ns
ME-8_4ns
API=44.2
1.0
.]u
.a
y[it
sn
et
in
de
zli
a
rm
on
0.8
0.6
700
450
500
550
600
I
1.0
ME-9_0.5ns
ME-9_1ns
ME-9_2ns
ME-9_4ns
API=47.8
.]u
.a
[y
ti
sn
et
in
de
zli
a
rm
on
0.8
0.6
0.4
650
wavelength[nm]
700
J
0.4
0.2
0.2
0.0
0.0
450
500
550
600
wavelength[nm]
650
700
450
500
550
600
650
700
wavelength[nm]
Figure 2: (A-J) Time-resolved Emission Spectra (TRES) of ten Middle Eastern oils measured over 450–700 nm wavelength range
with 405 nm excitation. The spectra were normalized at the wavelength of maximum fluorescence intensity (λmax). For the ease of
comparison the oils are listed in order of increasing API gravity.
Similarly to steady-state data the TRES show broad emission bands (Figure 2 A-J) devoid of any vibronic structure,
which indicates that the fluorescence emission originates from complex mixtures of fluorophores with overlapping
emission and absorption spectra.16,17 Two chief processes that govern the fluorescence dynamics of crude oils are energy
transfer and fluorescence quenching.16,17,19-21,23,31,32 An important point of note is that steady-state fluorescence emission
data is the average of the time-resolved emission spectra over the intensity decay of the sample and as a consequence
some information is irreversibly lost in the time averaging process. Because TRES are constructed from fluorescence
intensities sampled at discrete times during the decay of the sample, an obvious implication is that TRES provide more
insight into fluorescence dynamics then steady-state based measurements and thus should provide more information
about fluorescence quenching and energy transfer. In general, one would expect to find that the spectral changes in
TRES follow similar trends to those observed in steady-state spectra. Briefly, the position of the wavelength of
maximum fluorescence emission intensity (λmax) and spectral width (FWHM‡) are primarily influenced by the extent of
collisional energy transfer process, whereas fluorescence intensity is strongly influenced by quenching rates. However,
there was no attempt made to measure the absolute fluorescence intensities. Moreover, the normalization method used,
i.e. the individual TRES spectra were normalized at wavelength of the maximum fluorescence intensity, obscures the
relative difference between the fluorescence intensities for the different time gates.
120
120
A
2
r =0.81
B
100
100
s
n
5
.
0
sn 80
.50
xa
m
λ
−s 60
n4
xa
λm 40
80
M
H
W
F
-
60
s
n
4
M
H
W
F
40
2
r =0.79
20
20
0
0
25
30
35
40
API[deg]
45
50
25
30
35
40
45
50
API[deg]
Figure 3: Plot of API gravity for ten Middle Eastern oils versus: (A) the time gate dependent batochromic shift (λmax4ns-λmax0.5ns); (B)
the time gate dependent spectral width difference (FWHM4ns-FWHM0.5ns). The dashed lines represent the linear fits (r2 stands for the
correlation coefficient).
3.1.1.
Spectral correlations
The position of λmax is primarily governed by the extent of collisional energy transfer so it depends above all on the
nature, concentration, and population distribution of chromophores and fluorophores in the oils. In brief, heavy oils (low
‡
The acronym FWHM refers to the full spectral width measured at half maximum of fluorescence intensity
API gravity) tend to have more red-shifted fluorescence bands then lighter, usually more mature, oils (higher API
gravity). It stems from the fact that there are fewer fluorophores in light oils. Moreover, light oils contain relatively lower
populations of large, small energy band gap, aromatics as compared to heavy oils.16 Overall this leads to a reduction in
energy transfer rates and less red-shifted emission. The common feature of crude oils TRES is also a progressive shift to
the longer emission wavelength with increasing time gates (Figure 2 A-J). This time gate dependent batochromic shift is
symptomatic of higher contribution of the energy transfer produced emission at later time gates to the total emission and
the difference between λmax4ns and λmax0.5ns varies from ~100 nm for the heavy oils to <20 nm for the light oils. The value
of that batochromic shift depends on the concentration and on the specific distribution of fluorophores and is negatively
correlated with API gravity (Figure 3 A).
Similar argument can be advanced for time gate dependent spectral width difference (FWHM4ns-FWHM0.5ns) and API
gravity also shows negative correlation with FWHM4ns-FWHM0.5ns (Figure 3 B).
It was already established that the alkane fraction at 405 nm excitation wavelength acts as an inert solvent in which
fluorophores and quenchers are diluted.23,32
120
2
r =0.83
A
120
B
100
100
s
n
.5
0
sn 80
.50
ax
m
λ
−s 60
n4
xa
λm 40
80
M
H
W
F 60
s
n
4
M
H 40
W
F
20
2
r =0.76
20
0
0
50 55 60 65 70 75 80 85 90
corrected alkane[%]
50 55 60 65 70 75 80 85 90
corrected alkane[%]
Plot of corrected alkane concentration for ten Middle Eastern oils versus: (A) the time gate dependent batochromic shift
(λmax4ns-λmax0.5ns); (B) the time gate dependent spectral width difference (FWHM4ns-FWHM0.5ns). The dashed lines represent the linear
fits (r2 stands for the correlation coefficient).
Figure 4:
As such, the increase in alkane fraction causes a reduction in the collisional energy transfer rates. Therefore one would
expect to find that alkane fraction correlates negatively with λmax4ns-λmax0.5ns and FWHM -FWHM . This is indeed the
4ns
0.5ns
case as can be seen from the Figure 4 A, B.
Because fluorescence of crude oils derives from the presence of polycyclic aromatics, a natural implication is that
aromatic concentration should correlate with spectral parameters.
120
A
120
2
B
100
r =0.83
100
s
n
.5
0
sn 80
.50
xa
m
λ
−s 60
n4
xa
λm 40
80
M
H
W
F
-
60
s
n
4
M
H
W
F
2
r =0.76
40
20
20
0
0
10
15
20
25
30
35
40
corrected aromatics[%]
45
10
15
20
25
30
35
40
corrected aromatics[%]
45
Plot of corrected aromatic concentration for ten Middle Eastern oils versus: (A) the time gate dependent batochromic shift
(λmax4ns-λmax0.5ns); (B) the time gate dependent spectral width difference (FWHM4ns-FWHM0.5ns). The dashed lines represent the linear
fits (r2 stands for the correlation coefficient).
Figure 5:
There is a general trend towards higher values of λmax4ns-λmax0.5ns and FWHM4ns-FWHM0.5ns with increasing concentration
of aromatic compounds, although there is a significant degree of scatter around the trend line (Figure 5 A, B). This
general trend is due to increased rates of energy transfer.
However, the correlations between spectral parameters (λmax4ns-λmax0.5ns and FWHM4ns-FWHM0.5ns) and compositional
data, discussed above, while useful for general qualitative discussion, are rather not suitable for qualitative analysis.
3.1.2.
Contour plots
An alternative way to visualize TRES is by means of contour plots that represent the three-dimensional time gateemission wavelength-intensity data set as two-dimensional maps (time gate–emission wavelength) with isointensity
contours. This approach allows a general comparison between different crude oils on the basis of concentration and
chemical composition.
The compositional diversity of ME oils results in a range of TRES contour plots as seen in Figure 6 A-J. These contour
plots were constructed using the TRES data as in Figure 2, i.e. the individual TRES spectra were normalized at the
wavelength of maximum fluorescence intensity (λmax). With this method of normalization, the wavelength distribution of
fluorescence intensity reveals information about the extent of the energy transfer.
A) ME-10 24.7 API
B) ME-2 29.8 API
4.0
4.0
0.70
0.80
3.5
0.40
]s
n[
e
m
it
2.5
0.90
0.90
0.50
]s
n[
e
m
it
0.60
2.0
1.5
1.0
0.90
0.800.70
0.60 0.50
0.90
0.50
0.80 0.70
3.0
0.80
3.5
3.0
0.60
2.5
0.70
0.40
2.0
1.5
0.90
0.70
0.80
0.60
0.90
0.80 0.70
1.0
0.50
0.80
0.90
0.90
0.80 0.70
0.60 0.50
0.30
0.40
0.40
0.5
0.5
450
500
550
600
650
700
450
500
550
600
650
wavelength[nm]
wavelength[nm]
C) ME-3 30.8 API
D) ME-1 32.7 API
4.0
700
4.0
0.40
3.5
0.90
0.90
0.60
0.50
]s
n[
e
m
it
3.0
2.5
0.40
]s
n[
e
m
it
0.80
2.0
0.30
1.5
1.0
0.90
3.0
0.60
2.5
0.70
2.0
1.0
0.40
0.80
0.90
0.90
0.50
0.80 0.70
0.60
0.20
0.5
0.30
0.5
450
500
550
600
650
700
450
500
550
600
650
wavelength[nm]
wavelength[nm]
E) ME-4 39.2 API
F) ME-6 39.6 API
4.0
700
4.0
0.90
3.5
0.90 0.800.70
0.60
0.50
0.70
]s
n[
e
im
t
0.70
0.50
0.80
0.60
1.5
0.90
0.80
0.50
0.70
0.40
0.60
0.90
0.90
0.50
1.00
0.70
0.80
3.5
0.80
0.700.60
0.90
3.5
0.40 0.30
3.0
0.70
0.20
0.80
2.5
2.0
1.5
]s
n[
e
im
t
0.80 0.70
0.90
0.60
0.500.400.30 0.20
3.0
0.80
2.5
2.0
1.5
0.10
1.0
0.90
0.80
0.70
0.60
0.90
0.500.40 0.30
1.0
0.20
0.90
0.90
0.60
0.500.400.30 0.20
0.70
0.80
0.10
0.5
0.5
450
500
550
600
wavelength[nm]
Figure 6: Continued on the next page.
650
700
450
500
550
600
wavelength[nm]
650
700
G) ME-7 42.1 API
H) ME-5 44.1 API
4.0
4.0
3.5
0.80
0.90
0.70
0.60
0.50
0.40
0.300.20
3.5
0.10
0.90
0.80
0.70
0.60
0.50
0.400.30
0.80
0.20
0.10
]s
n[
e
m
ti
3.0
]s
n[
e
m
ti
2.5
2.0
1.5
1.0
3.0
0.90
2.5
2.0
1.5
0.30
0.20
0.90
0.80
0.70
0.60
0.50
0.40
0.10
0.50
0.40 0.30 0.20
0.60
0.90
0.80
0.70
1.0
0.10
1.00
0.5
0.5
450
500
550
600
650
700
450
]s
n[
e
m
ti
600
650
wavelength[nm]
I) ME-8 44.2 API
J) ME-9 47.8 API
700
4.0
0.90
0.80
0.70
0.60
0.50
0.40
0.30 0.20
3.5
0.10
3.0
]s
[n
e
m
ti
2.5
2.0
0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
3.0
2.5
2.0
1.5
1.5
1.0
550
wavelength[nm]
4.0
3.5
500
0.20
0.90
0.80
0.70
0.60
0.50
0.40
0.30
1.0
0.10
0.20
0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.10
0.5
0.5
450
500
550
600
wavelength[nm]
650
700
450
500
550
600
650
700
wavelength[nm]
(A-J) Contour TRES plots of ten Middle Eastern oils measured for 0.5-4 ns time gates over 450-700 nm wavelength range
with 405 nm excitation. For comparison purposes the oils were listed from A to J in order of increasing API gravity.
Figure 6:
The crude oil TRES contour plots show a general trend for shifting of the isointensity contours from long emission
wavelength to short emission wavelength with increasing maturity (Figure 6 A-J). This trend represents the gradual
decrease in energy transfer and quenching rates with increasing oil maturity.
The heaviest oil, ME-10 (Figure 6 A), which has relatively high concentration of polar and asphaltene compounds as
compared to the other ME oils, shows prominently different contour plot topology. For ME-10, the contour lines
representing intensity from 1.0 down to 0.9 stretch diagonally from ~450 nm to ~525 nm and from ~525 to ~625 nm
window. This broad distribution of relatively red-shifted fluorescence intensity, as related to the other ME oils, is
symptomatic of the large contribution of energy transfer excited fluorescence. The contour plot for oil ME-2 looks fairly
similar (Figure 6 B), however the 0.9 intensity lines span diagonally from ~450 nm to ~480 nm and from ~480 nm to
~575 nm, which indicates a small drop in energy transfer compared to ME-10. Oil ME-10 represents lower alkane and
higher aromatic concentrations resulting in elevated rates of energy transfer as compared to ME-2.
The bottom row (Figure 6 I, J) shows TRES contour plots for the two lightest, the most mature oils ME-8 and ME-9,
which have a relatively large concentration of alkane constituents (~80%) combined with a rather low aromatic
concentration (~20%). In each case, the contour lines representing the intensity down to 0.9 fall into ~450-465 nm
window. This drift of contour lines towards shorter emission wavelengths is a result of a drop in energy transfer in the
light oils.
Moving from the top of Figure 6 to the bottom one observes a progressive reorientation of isointensity contours from
longer to shorter emission wavelength. This change in contour plot topology is due to a gradual decrease of the energy
transfer rates with increasing oil maturity.
3.2.
Fluid inclusion case study
F3a
4.0
3.5
0.900.80
0.80
0.70
0.60 0.50
0.40
0.90
0.20
0.30
3.0
0.90 0.80
]s 2.5
[n
e
m
it 2.0
0.90
1.5
0.70
0.60
0.50
0.40
0.70 0.60
0.80
0.50
0.40
0.20
0.30
0.90
0.30
0.20
0.90
1.0
0.70 0.600.50
0.90
0.80
0.40
0.30
0.20
0.10
0.5
450
500
550
600
wavelength[nm]
650
700
Figure 7: TRES contour plot of HCFI F3a.
Figure 7 show TRES contour map for fluid inclusion F3a. The contour lines representing the 0.9 intensity lines fall into
~450-460 nm and ~475-510 nm window and overall the plot looks fairly similar to that of the oil ME-5 (compare Figure
7 and Figure 6 H). Our previous studies based on the intensity average fluorescence lifetime have identified inclusion
F3a as containing high alkane and low polar concentrations hydrocarbon fluid.32 Also the recovered values and
wavelength dependence of τ for F3a32 and ME-537 (~3-9 ns) are fairly similar. The above indicates that the oil entrapped
in the inclusion F3a comprises similar types and distributions of fluorophores as the oil ME-5.
4.
CONCLUSIONS
The gross composition metrics obtained from standard analysis correlate with spectral changes in TRES for a
geographically related set of crude petroleum oils. This suggests scope for developing accurate quantitative methods for
oils characterization using TRES. All the crude oil TRES contour plots show a general trend for isointensity lines to
move from long emission wavelength to short emission wavelength with increasing maturity. This trend represents the
gradual decrease in energy transfer rates with increasing oil maturity.
We have presented the first TRES collected from individual HCFI. TRES contour plots allow us to fingerprint the single
HCFI and to peg the data to bulk oil contour maps.
In summary, the normalization method used allows for qualitative analysis of a dominant process in crude oil
fluorescence, the energy transfer. However, it masks the relative differences in fluorescence intensities measured at
different time gates for the individual oils. Moreover, as with any form of normalization it obscures huge variations in
intensities observed for different crude oil samples. Currently we are investigating a different method of TRES
normalization by looking at the spectra with respect to their relative intensity i.e. 0.5 ns spectrum is normalized at λmax
and the traces at later time gates (1-4 ns) are scaled to it. This approach, which retains the relative intensities, should
provide the information about the extent of fluorescence quenching.
ACKNOWLEDGEMENTS
The authors wish to express gratitude to Mark Bastow and Petroleum Geochemistry Division of Fugro Robertson,
Llandudno, North Wales for providing the bulk crude oil samples along with associated compositional data. We also
would like to thank Stan Perry for supplying and carefully selecting the fluid inclusion samples, and Alfie Sherlock
(Department of Earth and Ocean Sciences, NUI, Galway) for producing the sample rock wafers. This work was assisted
by the National Center for Biomedical Engineering Science as part of Higher Education Authority’s Programme for
Research in Third Level Institutions.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
E. Roedder,
, Reviews in mineralogy, Vol. , Mineralogical Society of America, Washington,
1984.
D. Emery, A. Robbinson, Inorganic geochemistry: Applications to petroleum geology, Blackwell Scientific
Publications, Oxford, London, Edinburgh, 1993.
R.K. Mclimans, "The application of fluid inclusions to migration of oil and diagenesis in petroleum reservoirs,"
Applied Geochemistry, , pp. 585-603, 1987.
R.C. Burruss, Petroleum fluid inclusions, an introduction, in Fluid inclusions: Analysis and interpretation, pp.
159-169, Mineralogical Association of Canada, Vancouver, 2003.
I.A. Munz, "Petroleum inclusions in sedimentary basins: systematics, analytical methods and applications,"
Lithos, , pp. 195-212, 2001.
R.C. Burruss, Crushing cell, capillary column gas chromatography of petroleum fluid inclusions: method and
application to petroleum source beds, reservoirs, and low temperature hydrothermal ores, in Fluid inclusion
research, pp. 59, University of Michigan Press, Chicago, 1987.
S.C. George, F.W. Krieger, P.J. Eadington, R.A. Quezada, P.F. Greenwood, L.I. Eisenberg, P.J. Hamilton, M.A.
Wilson, "Geochemical comparison of oil-bearing fluid inclusions and produced oil from the Toro sandstone,
Papua New Guinea," Organic Geochemistry, , pp. 155-173, 1997.
P.F. Greenwood, S.C. George, M.A. Wilson, K.J. Hall, "A new apparatus for laser micropyrolysis-gas
chromatography/mass spectrometry," Journal of Analytical and Applied Pyrolysis, , pp. 101-118, 1996.
R. Li, J. Parnell, "In situ microanalysis of petroleum inclusions by Time of Flight-Secondary Ion Mass
Spectrometry as an indicator of evolving oil chemistry: a pilot study in the Bohai Basin, China," Journal of
Geochemical Exploration,
, pp. 377-384, 2003.
N. Guilhaumou, N. Szydlowski, B. Pradier, "Characterization of hydrocarbon fluid inclusions by infra-red and
fluorescence microscpectrometry," Mineralogical Magazine, , pp. 311-324, 1990.
J. Pironon, O. Barres, "Semi-quantitative FT-IR microanalysis limits: Evidence from synthetic hydrocarbon
fluid inclusions in sylvite," Geochimica et Cosmochimica Acta, , pp. 509-518, 1990.
J. Pironon, J. Sawatzki, J. Dubessy, "Letter: NIR FT-Raman microspectroscopy of fluid inclusions:
Comparisons with VIS Raman and FT-IR microspectroscopies," Geochimica et Cosmochimica Acta, , pp.
3885-3891, 1991.
S.C. George, T.E. Ruble, A. Dutkiewicz, P.J. Eadington, "Assessing the maturity of oil trapped in fluid
inclusions using molecular geochemistry data and visually-determined fluorescence colours," Applied
Geochemistry, , pp. 451-473, 2001.
R.C. Burruss, Hydrocarbon fluid inclusions in studies of sedimentary diagenesis, in Fluid inclusions:
Applications to petrology, pp. 138-156, Mineralogical Association of Canada, Calgary, 1981.
J.R. Lakowicz, ed. Topics in Fluorescence Spectroscopy, Techniques, Vol. , Plenum Press, New York and
London, 1991.
O.C. Mullins, E.Y. Sheu, eds., Structure and dynamics of asphaltenes, Plenum Press, New York, 1998.
T.D. Downare, O.C. Mullins, "Visible and Near-Infrared Fluorescence of crude oils," Applied Spectroscopy, ,
pp. 754-764, 1995.
C.Y. Ralston, X. Wu, O.C. Mullins, "Quantum yields of crude oils," Applied Spectroscopy, , pp. 1563-1568,
1996.
X. Wang, O.C. Mullins, "Fluorescence lifetime studies of crude oils," Applied Spectroscopy, , pp. 977-984,
1994.
A.G. Ryder, T.J. Glynn, M. Feely, A.J.G. Barwise, "Characterization of crude oils using fluorescence lifetime
data," Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, , pp. 1025-1037, 2002.
Fluid inclusions
12
2
55
26
38
78-79
54
54
55
13.
14.
15.
16.
17.
18.
19.
20.
16
1
49
50
48
58
21.
22.
23.
24.
25.
A.G. Ryder, "Quantitative analysis of crude oils by fluorescence lifetime and steady state measurements using
380nm excitation," Applied Spectroscopy, , pp. 107-116, 2002.
A.G. Ryder, T.J. Glynn, M. Feely, "Influence of chemical composition on the fluorescence lifetimes of crude
petroleum oils," Proceedings SPIE,
, pp. 1188-1195, 2003.
A.G. Ryder, "Time-resolved fluorescence spectroscopic study of crude petroleum oils: influence of chemical
composition," Applied Spectroscopy, , pp. 613-623, 2004.
H.W. Hagemann, A. Hollerbach, "The fluorescence behaviour of crude oils with respect to their thermal
maturation and degradation," Advances in Organic Geochemistry, , pp. 473-480, 1986.
N.H. Oxtoby, "Comments on: Assessing the maturity of oil trapped in fluid inclusions using molecular
geochemistry data and visually-determined fluorescence colours," Applied Geochemistry, , pp. 1371-1374,
2002.
S.C. George, T.E. Ruble, A. Dutkiewicz, P.J. Eadington, "Reply to comment by Oxtoby on "Assessing the
maturity of oil trapped in fluid inclusions using molecular geochemistry data and visually-determined
fluorescence colours," Applied Geochemistry, , pp. 1375-1378, 2002.
J.R. Lakowicz, Principles of Fluorescence Spectroscopy. 2 ed, Kluwer Academic/Plenum Publishers, New
York, Boston, Dordrecht, London, Moscow, 1999.
R.M. Measures, W.R. Houston, D.G. Stephenson, "Laser induced fluorescence decay spectra- a new form of
environmental signature," Optical Engineering, , pp. 494-501, 1974.
D.M. Rayner, A.G. Szabo, "Time-resolved laser fluorosensors: laboratory study of their potential in remote
characterization of oil," Applied Optics, , pp. 1624-1630, 1978.
M.E. Abu-Zeid, K.S. Bhatia, M.A. Marafi, Y.Y. Makdisi, M.F. Amer, "Measurement of fluorescence decay of
crude oil: a potential technique to identify oil slicks," Environmental Polllution, , pp. 197-207, 1987.
A.G. Ryder, "Assesing the maturity of crude petroleum oils using total synchronous fluorescence scan spectra,"
Journal of Fluorescence, , pp. 99-104, 2004.
A.G. Ryder, M.A. Przyjalgowski, M. Feely, B. Szczupak, T.J. Glynn, "Time-resolved fluorescence
microspectroscopy for characterizing crude oils in bulk and hydrocarbon-bearing fluid inclusions," Applied
Spectroscopy, , pp. 1106-1115, 2004.
E. Hegazi, A. Hamdan, J. Mastromarino, "New Approach for Spectral Characterization of Crude Oil Using
Time-resolved Fluorescence Spectra," Applied Spectroscopy, , pp. 202-207, 2001.
E. Hegazi, A. Hamdan, "Estimation of crude oil grade using time-resolved fluorescence spectra," Talanta, ,
pp. 989-995, 2002.
A.G. Ryder, T.J. Glynn, M.A. Przyjalgowski, B. Szczupak, "A compact violet diode laser based fluorescence
lifetime microscope," Journal of Fluorescence, , pp. 177-180, 2002.
W.E. Schneider, D.G. Goebel, "Standards for calibration of optical radiation measurement systems," Laser
focus/Electro optics, , pp. 12-20, 1984.
M.A. Przyjalgowski, A.G. Ryder, M. Feely, T.J. Glynn, "paper in preparation," 2005.
56
4876
58
10
17
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
17
13
17
46
14
58
55
56
12
9