oil type as well as excitation wavelength; quantum yields are lower for high fluoropbore concentrations (heavy crude oils) and for longer wave- length excitation.
Visible and Near-Infrared Fluorescence of Crude Oils* TAGGART D. D O W N A R E and OLIVER C. M U L L I N S t Schlumberger-Doll Research, Old Quarry Road, Ridgefteld, Connecticut 06877
Fluorescence emission spectra and absolute quantum yields have been measured for ten diverse crude oils at various concentrations over a broad range of excitation and emission wavelengths in the visible and the nearinfrared. Energy transfer produces large red shifts and large widths in the fluorescence emission spectra for shorter wavelength excitation particularly for heavier crude oils. However, the effects of energy transfer are nearly absent for near-infrared excitation; all crude oils exhibit nearly the same emission spectra for long wavelength excitation. In addition, the fraction of emission resulting from collisional energy transfer relative to nascent emission is almost independent of oil type; it is governed by quantum yield characteristics. Absolute fluorescence quantum yields of ten crude oils (and three rhodamine dyes for validation) were measured with respect to scattering of latex microspheres in distilled water. Fluorescence quantum yields vary systematically with crude oil type as well as excitation wavelength; quantum yields are lower for high fluoropbore concentrations (heavy crude oils) and for longer wavelength excitation. Stern-Volmer analyses of the quantum yields indicate that simple models apply and show the relative quenching rates for different excitation wavelengths. Index Headings: Crude oil; Petroleum; Fluorescence; Quenching; Energy transfer; Quantum yield; Spectra; Chromophore; Fluorophore; Visible, Near-infrared.
INTRODUCTION Fluorescence spectra and absolute quantum yields provide a wealth of information for a variety of purposes. Luminescence properties of chemical systems can be very revealing; specifically, fluorescence studies of crude oils have provided valuable insight into their fundamental and dynamic properties. In addition, fluorescence properties of crude oils and related organic matter have been used in a variety of applications. Evolution of organic material in source rock can be monitored by fluorescence. 1 The fluorescence of crude oils is widely used to find oil shows during drilling of oil wells. 2 Our interest is to uncover the systematics of the optical properties of crude oils and to develop a better understanding of their dynamic behavior. For instance, the electronic absorption edge for all crude oils 3 and asphaltenes 4 exhibits the Urbach tail (exponential profile) and is related to the population distribution of chromophores. Optical measurements on crude oils have led to the development and commercial use of an optical analyzer in the oil field for the downhole environment? The temperature dependence of crude oil fluorescence has been determined and related to thermal quenching and, to a lesser extent, complex formation. 6 Complexes are important for heavy crude oils and asphaltenes even at very low concentrations. 7 Fluorescence lifetimes, spectra, and Received 9 September 1994; accepted 15 February 1995. * Based in part on the Masters Thesis of T.D.D., Dept. of Physics and Applied Optics, Rose-Hulman Institute of Technology, Terre Haute, IN 47803. I" Author to whom correspondence should be sent.
754
Volume 49, Number 6, 1995
relative quantum yields of crude oils have been shown to vary systematically? Concentration-dependent effects of energy transfer and quenching reduce fluorescence lifetimes for high concentrations and heavy crude oils? Also, collisional effects on fluorescence lifetimes were shown to obey first-order kinetics, and collisional processes were found to be efficient for energy transfer and quenching, s Here, we investigate the evolution of electronic excitation energy in crude oils. We establish the variation of fluorescence emission spectra of crude oils with concentration, illustrating the effects of collisional energy transfer. The extent of collisional energy transfer varies dramatically with excitation wavelength from the UV to the near-infrared, but is independent of crude oil type. We also determine absolute quantum yields for different concentrations of crude oils to establish the significance of nonradiative decay and collisional quenching. Nonradiarive processes are dominant even in the dilute limit, indicating that only a small fraction of chromophores in crude oils are fluorescent (within our spectral range); this fraction decreases with increasing excitation wavelength. Furthermore, quantum yields of all dilute crude oils exhibit nearly the same behavior with excitation wavelength. The quantum yield systematics of dilute crude oils account for the trends found in collisional energy transfer. Collisional quenching becomes an important process at higher concentrations of chromophores and is particularly large for neat, heavy crude oils.
EXPERIMENTAL Ten crude oils, from light to heavy, were obtained from different oil fields around the world. Crude oil American Petroleum Institute (API) gravities range from 13 to 43 and cutoff wavelengths range from 493 to 1453 nm. The cutoff wavelength is defined 3 as the wavelength where the electronic absorbance equals 3 for a 2-ram pathlength; thus, heavier crude oils have larger cutoff wavelengths. Dilution of the crude oil samples was performed with spectrophotometric-grade benzene from Aldrich Chemical Company. All samples were measured in the same 10-mm-pathlength UV-grade quartz cell from Stama Cells to reduce possible problems associated with cell imperfections or nonuniformities. Rhodamine B (610), rhodamine 6G (590), and rhodamine 101 (640) were used without purification from Exciton. H P L C - g r a d e methyl alcohol from Aldrich Chemical Company was used as the solvent for the dye solutions. Oxygen quenching of fluorescence can be significant. 6 Deaeration of the sample solutions to remove oxygen was performed with the quartz cell with a gas chromatography (GC) septum stopper and two GC needles to inject and vent N2 gas. Comparison of dye and crude oil samples purged with the N 2 gas for 0.25 h to those that were not
0003-7028/95/4906-075452,00/0 ©
1995 Societyfor AppliedSpectroscopy
APPLIED SPECTROSCOPY
deaerated revealed that there is not a noticeable change in the spectra and there is less than 3% change in the measured power levels for quantum yield calculations. The lack of oxygen quenching is in part due to the short fluorescent lifetimes 8 and dominance of other quenching mechanisms in crude oils. Various methods can be used for measuring quantum yields: magnesium oxide as a standard (Weber and Teale technique), 9-t~ suspensions of scatterers as standards, comparison with compounds of known (large) quantum y i e l d s , t2-14 calorimetric methods (for measurements of large quantum yield), tS,t6 or absolute evaluation of the particular geometry. ~7 The reliability of comparative techniques requires close attention to excitation wavelength, concentration, ts.t9 temperature,20 solvents,2~ emission spectra, absorbance, reabsorbance, and refractive index of the solutions. Here, light scattering suspensions made with uniform polystyrene latex microspheres from Duke Scientific Corporation, with mean diameters of 0.107, 0.494, 0.596, and 10.2 urn, were used as standards for determination of quantum yields. The scattering microspheres did not exhibit fluorescence for the excitation wavelengths used in this work. As expected, a dilute 10.2-urn microsphere suspension was the only microsphere suspension to exhibit a uniform absorbance profile across the wavelength range used here. Distilled water was used to dilute the microsphere suspensions to match approximately the apparent absorption of the scattering suspensions and crude oil samples. The particles were easily suspended by gently rolling or swirling the sample cell. The larger particles (specifically the 10.2 #m) gradually settled to the bottom of the cuvette over a period of hours; therefore, the sample cell was rolled or swirled between measurements. The measurements were found to be repeatable with this method. All absorption spectra and absorption measurements were obtained on a Cary 5 UV-Vis near-IR spectrophotometer. All fluorescence spectra shown for 300-, 350-, 400-, and 470-nm excitation were obtained with a PTI LS- 100 fluorescence spectrophotometer. The PTI was used in front-surface mode for the neat crude oils since this mode is compatible with the high optical density (OD) of the neat crude oils. The transmission mode of the PTI was used for the dilute ( < 0 . 1 0 D ) crude oil spectra. The PTI cannot accurately measure spectra past ~750 nm because of limitations of the photomultiplier tube (PMT) sensitivity; therefore, an alternative system must be used for the near-infrared spectra. All fluorescence spectra shown for 543-, 674-, 790-, and 838-nm excitation were obtained with an EG&G spectrograph system (Acton Research Corporation 0.275m-focal-length monochromator/spectrograph) equipped with a 512 × 512 charge-coupled device (CCD) detector (Model 1530-P) and OMA-Vision controller board connection to a Gateway 2000 4DX2-66V microcomputer, as depicted schematically in Fig. 1. The 150-grooves/mm grating, blazed at 500 nm, was used for all spectra obtained with this system. The EG&G spectrograph is basically an emission spectrophotometer where the excitation source must be supplied externally. The externally supplied sources were a 0.3-mW Melles Griot 543-nm green H e - N e laser (Model 05SGR871), a 5-mW Melles
AI 543 nm Green lteNe 790 nm La.~erDiode
A3 rl
M1, • ]M3 • 674 nm Laser Diode M~
838 nm Laser Diode
I"1
"%" M4
| A4
~
A2
~P1
ASi Detector
(b)
Bifurcated Fiber Optic Bundle
L1
FibBeur~dOlP~C:
E L2O i
'\'/~
OL2 i i
I
EG & 13 Spectrograph System .........................
,
EG & G Spnctrograph System, Monitoring ', Detector ,. . . . . . . . . . . . . . . . . . . . . . . . . . . .
FIG. 1. Schematic diagram of the system used for measurements of fluorescence spectra and absolute quantum yields for (a) concentrated and (b) dilute solutions.
Griot 674-nm laser diode (Model 06DLL601), a 30-mW Melles Griot 790-nm laser diode (Model 06DLL401), and a 40-mW Melles Griot 838-nm laser diode (Model 06DLL503). Laser line filters (F1 = 543df10, F2 = 674dfl 0, F3 = 790dfl 0, and F4 = 838dfl 0) were obtained from Omega Optical. Newport Corporation optical power meters (Model 838) with a 1-cm ~ silicon detector (Model 818UV) and OD 3 attenuator (Model 883UV) were used here. Apertures and optical mounts were also obtained from Newport Corporation. Fiber-optic bundles used here (high O H - silica-silica fibers) were obtained from Electro Fiber Optics Corporation. The high O H - silica-silica fibers were used to minimize fiber-optic fluorescence. 22 An analysis was performed of the EG&G spectrograph throughput with a Newport (Model 780) 50-W 3200 K tungsten-halogen source placed at the sample location S1 in Fig. 1. The actual spectrum obtained deviates from the true blackbody radiation curve of the tungsten-halogen source largely because of the spectral response curve of the silicon in the CCD array and also the diffraction efficiency of the 150-grooves/mm grating. All spectra obtained with the EG&G were corrected by this spectrograph transmission function. We compared fluorescence spectra obtained with different fluorescence spectrophotometers to determine the accuracy of the throughput corrections. Fluorescence spectra were obtained with both the EG&G and the LS100 spectrophotometers at 543-nm excitation with a dilute solution of Sales crude oil. The spectra, including corrections, were virtually identical, indicating that our correction performed on the EG&G spectra (obtained with a CCD camera) compares extremely well with that for the LS-100 spectra (which are corrected by the PMT response curve). Spectra of other crude oils were compared with the use of the EG&G and LS-100 spectroAPPLIED SPECTROSCOPY
755
)0 °
1
0.8 0.6 =
0.4
=
0.2
k
l}l~
0
0°
N
--
0.2
0.4 © c_ Z 0.6 ~0.8 ~-
L
~ 0.107 urn Microspheres ] - - - - Ideal Scatterer .... Brunei Crude Oil . . . . . . . Acoustics Crude Oil ...................
I
i "1 .......................
hi
I
FIG. 2. Angular emission profiles of scattering from a concentrated solution of 0.107-urn microspheres and of fluorescence from two neat crude oils, which are nearly identical to the ideal Lambertian profile
(cos 0). photometers and exhibited similar agreement. Therefore, it appears that instrumental effects have been removed from our spectra and that all of our spectra collected with different systems are directly comparable. A B S O L U T E Q U A N T U M YIELDS Figure 1 shows the arrangement used to obtain the necessary information to calculate fluorescence quantum yields of (a) concentrated and (b) dilute solutions using concentrated and dilute scattering solution standards, respectively. Scattering standard performance was verified by plotting the apparent optical density of the solutions against X-4. For the 0.107-#m microspheres, this plot gave a straight line which extrapolated to zero at infinite wavelength, indicating strictly dipolar scattering without appreciable absorption. Figure 2 shows the angular emission profiles for a concentrated solution of 0.107-#m microspheres and two neat crude oils compared to the ideal Lambertian profile (cos 0) with the use of linearly polarized 543-nm excitation; nearly an ideal Lambertian scattering profile is achieved. Because the spheres show no absorption, we can measure quantum yields for concentrated solutions for each of our source wavelengths by
~" \E]\~]\
Rd
. (1 + R,).
(13
Subscripts u and s refer to the sample of unknown quantum yield and scattered light, respectively. ~u is the quantum yield of the concentrated sample; P is the measured power including baseline subtraction, detector response correction, and normalization to the source power; n is the refractive index; Ru is the total corrected area under the unknown fluorescence spectrum; and Rus is the corrected area under the unknown fluorescence spectrum over the excitation wavelength band. The purpose of this term is to remove the effects of scattered source light in the determination of the fluorescence quantum yields. 756
Volume 49, Number 6, 1995
The exciting sources were linearly polarized vertical to the sample (perpendicular to the plane containing the incident and detected light rays). The emission profiles shown in Fig. 2 are Lambertian and randomly polarized; therefore, no polarization correction factor was included in Eq. 1. Self-absorption effects are accounted for with the factor (1 + R,). With the assumption that the fluorescence travels the same distance in the sample as the excitation light (close to our experimental geometry), the ratio of fluorescence produced in the sample divided by the fluorescence which exits the sample is given by (1 + R,) where R , is the ratio of extinction coefficients at the fluorescence vs. excitation wavelengths, respectively. For each of the excitation wavelengths, the mean red shift can be determined from the spectra shown in Fig. 3; values for the excitation wavelength and mean red shift of the fluorescence are 543 and 87 nm; 764 and 66 nm; 790 and 55 nm; and 840 and 40 nm. With the use of the universal exponential decay constant for crude oil absorption spectra, 3 2162 c m - 1, the values of Ra can be obtained for each excitation wavelength. The values of R , are 0.31 for 543 nm, 0.54 for 674 nm, 0.68 for 790 nm, and 0.78 for 838 am. The quantum yields of the dilute crude oil solutions at 543 nm are referenced against dilute scattering and dye solutions. The optical densities (or, for nonscattering solutions, the absorbances) were kept below 0.1 for a 10ram pathlength at the excitation wavelengths in order to minimize self-absorption effects and concentration quenching in the dilute crude oils. Further dilutions were found to yield the same absolute quantum yields. Selfabsorption is minimal in dilute solutions of crude oils because of low absorbances and decreasing electronic absorption at longer (fluorescence) wavelengths. 3 At 543 nm, a scattering solution of 0.107-um microspheres was used as our reference standard. --
R u
s
~. Kl~]\l(X.)/\p,/\n~/\ =
A(X) is the absorbance (or attenuation from scattering) of the solution at the exciting wavelength X, and I(X) is the relative intensity at the exciting wavelength X. Subscripts u and s refer to the sample of unknown quantum yield and scattering solution, respectively. K is a correction factor accounting for differences in the angular distribution between the scattered light (dipolar) and the fluorescence (isotropic). The ideal theoretical value for K is 1.5 for the source light polarized perpendicular to the plane of incidence and detection. Essentially, the Rayleigh scattered light is directed along two spatial directions only, while the fluorescence is emitted along all three; this gives K = 1.5. A corrected value for K of 1.43 was found from an analysis of the experimental setup by including factors such as actual source and detector sizes. The ideal theoretical value for K is 0.75 when unpolarized light is used. t7 Quantum yields of crude oil solutions at several additional wavelengths (543,674, 790, and 838 nm) were obtained. The small microspheres show the X-4 dependence; thus, we reference quantum yields to large spheres (10.2 ~m) in order to remove the wavelength dependence. The large spheres yield a different angular distribution of
scattering. Therefore, for the large spheres of a given OD, Eq. 2 was multiplied by a geometric factor accounting for the difference in angular distribution of the scattered light from small and large spheres in our light collection geometry; the factor was experimentally determined to be 0.63. For validation, fluorescence quantum yields for several dilute dye solutions were determined by using the 0.107#m microspheres as the scattering standard. All dye quantum yields were measured at 20°C, in methanol with 543nm excitation. Rhodamine 6G is found to have a quantum yield of 0.86, which is close to a reported value of 0.8. 's For rhodamine 101, we obtain a quantum yield of 0.88, which compares well with a value of 0.82 found in the literature. 2° The results found in this work are close (within 7%) to values determined by other work with different methods under similar conditions (excitation wavelength, solvent, temperature, concentration), t~ Variations of 10% in quantum yields are not unusual. ~6Rhodamine B is found to have a quantum yield of 0.64; we did not find literature values for identical conditions. For the same solvent, but much different source wavelength, rhodamine B is estimated to have a quantum yield of 0.55. 6 R E S U L T S AND D I S C U S S I O N Several competing processes occur with the deposition of electronic excitation energy in crude oils. These processes of interest here are indicated in Eqs. 3-10. Rate A + hv ~ A*
A* ~ A + h,' A* + B ~ B* + A B* ~ B + h," A* + Q ~ A + Q + `5 B* + Q ~ B + Q + `5
A* ~ A + A B* ~ B + ,5
Process
0ight absorption) (3) (fluorescence emission) (4) [B]kE (collisional energy transfer) (5) k'Fo (fluorescence emission) (6) [Q]k o (collisional quenching) (7) [Q']k'a (collisional quenching) (8) kNR (nonradiative decay) (9) kNR (nonradiative decay) (10) kFo
where A is a fluorophore, h is Planck's constant, v is the frequency of light, `5 is heat, * represents an excited electronic state, and B represents a fluorophore with a longer emission wavelength than A. Q and Q' are chromophores which quench A and B, respectively, kvo and k'Fo are the intrinsic fluorescence rates, kE is the energy transfer rate, and kQ and k'Q are the quenching rates. We define collisional energy transfer as the process in which an excited
molecule gives up its energy to a larger molecule which then radiates, giving a shifted fluorescence spectrum. The processes ofEqs. 9 and 10, depict nonradiative decay and refer to decay mechanisms involving collisions with the solvent molecules and intramolecular quenching. Crude oils also contain significant quantities of chromophores which are not fluorophores. By chromophore we mean a molecule which absorbs light in the wavelength range of interest. Fluorophores obviously are fluorescent; all fluorophores are chromophores, but not all chromophores are fluorophores. Nonfluorescent chromophores of crude oils such as paramagnetic metalloporphyrins cause even dilute solutions of crude oils to have small quantum yields and are thus of great importance. They also can act as the quenchers shown in Eqs. 7 and 8, and these effects must be considered for concentrated solutions of crude oils as well. Here, we are concerned with the evolution of electronic excitation energy of fluorophores, so we do not explicitly depict the excitation of nonfluorescent chromophores; nevertheless, this process does limit absolute quantum yields of crude oils. Figure 3 shows the fluorescence emission spectra often crude oils for many excitation wavelengths (300, 350, 400, 470, 543, 674, 790, and 838 nm). The left side of the figure shows spectra for neat crude oil solutions, while the right side shows spectra for crude oils in the dilute limit. The crude oils in all figures are listed in order of cutoff wavelength (essentially lighter to heavier). The excitation source peak is clearly seen in the spectra obtained with the EG&G CCD, because of scattered light from the sample. The fluorescence emission spectra show that there are a variety of fluorophores involved in the production of fluorescence from the crude oils. There is a large difference between the neat and dilute crude oil spectra for short wavelength excitations; thus collisional energy transfer is very important with high concentrations. Also, at short excitation wavelengths, there are significant variations in the spectra for different crude oils, particularly for the neat samples. Therefore, the particular distribution of fluorophores and their concentration strongly affects the fluorescence spectral profiles. The difference between spectra of neat and dilute crude oils decreases with increasing excitation wavelength. Also, the differences among spectra of different crude oils decrease with increasing excitation wavelength, although differences in the magnitude of the quantum yields persist, as will be discussed. Evidently, collisional energy transfer (with subsequent fluorescence emission) becomes less important for longer wavelength excitations even for very high concentrations associated with heavy crude oils. Undoubtedly, relevant collisions take place between chromophores; however, these collisions do not result in energy transfer (with emission). Quenching, not energy transfer, apparently dominates the collisional processes. Thus, it appears that molecular properties, not changes in collisional processes, dominate the observed trends. Figure 4 is a comparison of spectra for neat and dilute --..)
FIG. 3. (a-p) Fluorescence emission spectra for ten crude oils (listed in order of increasing coloration) for neat and dilute samples for a series of excitation wavelengths from the ultraviolet to the near-infrared. Large collisional energy transfer produces differences between spectra for neat and dilute solutions with short wavelength excitation. Collisional energy transfer is nearly absent for long wavelength excitation where the spectra for neat and dilute solutions are nearly identical.
APPLIED SPECTROSCOPY
757
600
. . . .
'
/
'
ex = 300 nm ~RFT05-04 __ _ Bmnei -- -- -- Vixbt, rg ...... Escravos
~ (a) Neat
500
/1
//
--
"~
x~
"~
~
100 ~ / , " / . / .
.
.
.-r-',-"r
380
.
430
, ,/.-~,
,,
","~-"r"
t'
(nm~ "~ ex __
600
-
',:, // '/,
--
~'"
' " "
_
,
'
430
480
~
530
1400 ~ _ _ . . . _ . _ _ . .
~ . ~
~cravos
--
Nonl~ Sea
.
- -.
400
....
"~,\
•
\
570
,
. . . .
I
-
.
.
N
; \
"
~36
-
--
~i
, ""
,
"
~
--
..... ........
~X
. "\
'
600
650
Wavelength (nm)
758
Volume 49, Number 6, 1995
Acoustics NorthSea
BG5 Sales UG8
~.:,:..:. #86'"~*',80
'
I
i
'
'
'
1
250
...... • .....
2OO
~, . . . \\~'''"
.
.
.
.
.
.
-
-
gmo 50 .-
2 ~0 1.
/~u
~'~,~*~..-~ i
L
i
i
.
i
]
500
.
z ? . ~ ... .
J
i
.
570
640
Wavelength
2 5 0 ~. , < , < . ~ .... P/ . ..>,~ Q .
05.04
;~O00 g
~-
E..F
North Sea
Escravos
Acoustics NaN, Sea ..... BG5 - - . . . . Sales . UG8 •.......... Vex141
--
\'~: ,,
150
vi~,,,~
RFT 05-04
_ Brunei -- -- -- Vixburg
": r..,
~
Dilute _ _
."'
7
(nm~~
, .... (h) Dilute
" ' " ~'- ' ' '
.
, el", ~' ' f - - - 2 . C
•.
~ ,' ~
e× = 4 7 0 n m _
-- -
-- Vixburg
...... -- . .
~
\',
', .
.
.
.
780
- __
RFT 05-04
......
"~.\.
-- "--
i
710
Brunei ~=vo,
Acoustics --
North
Sea
BG5
Sales UG 8
g 100 H . . . . . . . . . . .
ii"--.. " \ 550
Escravos
...... -- - -..... ..... ........
L,7
~
. .", "~ 3 0 0 ~[ ~ '
.~
~.- ~ .......... ...
0[
;00
580
........... Tex 141
(0
Acousiies
-
,
'-- ~ ' m T . . . - . ~ _ - ' ~ , , f
......
~04 ' '5%
350
Brunei
-
...... N x
,,
530
R.Vr O5.O4 - - Bmnei -- -- -- Vixburg
= 470 nm
r~ -
/
400
1 7 IO
i
N
'
~1000
Sea
BG5 Sales UG8 Tex 141
- -
"
%0
680
~
~g)
//
L .... I''' (d) Dilute
7=50 7_'5,. p.::2~:::.b~.,~.~.
NonhSeaBG5
(nm~L '~ ex
.
" Neat
"
"~ 1 5 0 0
Acoustics --NaN,
480
100 ~
141
ACoustiCs
640
.
-
~ ~"7'
Sales . . . . . . . . UG 8 ........... Tex 141
N
'
430
~
"~ 2 0 0
- - Bmnei -- Vixburg
"~'-"~":~'..%'-~" '
500
. . . .
'Escravos
--
......
Wavelength 2000
Vixburg
Wavelength (nm~. ex = 400 nm
..... X \
-
,
" ,'~
;~250
200 I0
t ....
150 7
630
- ..... ----
300 .....
Aex = 400 nm _ . . 1 (e) Neat - rWT05-04
• ,ooo r ...-.. " "" -
380
.
580 (nm).
-
"''',
330
. . . . . Sales ........ U G S
1200
600
--
Wavelength (nrn~ ex = 350 nm
- - v~x~,o,~
- -
800
Bru.ei
..... ..... ........ ...........
o v " ~ c : , ,~
,
580
......
.
~ . 2 . 2 £ _ ~ . ~ . . ~", ~ .
Wavelength
~
......
"',.\
""'
~0
\
150
Brunei
Tex
,
~200]
05-04
_
---
400
200-
--
......
= 350 nm
_
-
~,~
~
~
\
RUt 05.04
\
I
.....
530
' '
(c) Neat
/
'1"
480
~ ....
\
e , lOOO "~
(b) Dilute
.
"'r"l-.Y
Wavelength 1200
~
300 nm
IWT
'"
0
330
~, e x =
, ....
"--
North Sea BG 5 Sales UG 8 Tex 141
300
....
250
. . . . . . A~ous,ir-~ \ ,~
300
II
700
"•50
500
,
I
550
":,
