pubs.acs.org/Langmuir © 2009 American Chemical Society
Langmuir Films of Petroleum at the Air-Water Interface Vinı´ cius C. C. Vieira,† Divinomar Severino,*,† Osvaldo N. Oliveira, Jr.,‡ Felippe J. Pavinatto,‡ Maria E. D. Zaniquelli,§ Ana Paula Ramos,§ and Maurı´ cio S. Baptista† †
Instituto de Quı´mica, Universidade de S~ ao Paulo, 05315-970 S~ ao Paulo, SP, Brazil, ‡Instituto de Fı´sica de S~ ao Carlos, Universidade de S~ ao Paulo, 13560-970 S~ ao Carlos, SP, Brazil, and §Faculdade de Filosofia Ci^ encias e Letras de Ribeir~ ao Preto, Departamento de Quı´mica, Universidade de S~ ao Paulo, 14040-901 - Ribeir~ ao Preto, SP, Brazil Received May 19, 2009. Revised Manuscript Received August 7, 2009
Understanding the behavior of petroleum films at the air/water interface is crucial for dealing with oil slicks and reducing the damages to the environment, which has normally been attempted with studies of Langmuir films made of fractions of petroleum. However, the properties of films from whole petroleum samples may differ considerably from those of individual fractions. Using surface pressure and surface potential measurements and Brewster angle and fluorescence microscopy, we show that petroleum forms a nonhomogeneous Langmuir film at the air-water interface. The surface pressure isotherms for petroleum Langmuir films exhibit gas (G), liquid-expanded (LE), and liquidcondensed phases, with almost no hysteresis in the compression-decompression cycles. Domains formed upon compression from the G to the LE phase were accompanied by an increase in fluorescence intensity with excitation at 400-440 nm owing to an increase in the surface density of the chromophores in the petroleum film. The surface pressure and the fluorescence microscopy data pointed to self-assembling domains into a pseudophase in thermodynamic equilibrium with other less emitting petroleum components. This hypothesis was supported by Brewster angle microscopy images, whereby the appearance of water domains even at high surface pressures confirms the tendency of petroleum to stabilize emulsion systems. The results presented here suggest that, for understanding the interaction with water, it may be more appropriate to use the whole petroleum samples rather than its fractions.
Introduction The behavior of oil slicks in lakes, rivers, and sea depends on evaporation, dissolution, biodegradation, photooxidation, and interactions between petroleum, soil, sediment, and water.1,2 The combination of these processes is known as weathering, which causes changes in the chemical composition and physical properties of oil slicks.3 Such processes are strongly affected by environmental conditions at the site of the spill and by aging, which can last days, months, or even years. Oil spreading involves expansion of a thin layer of petroleum on water, the understanding of which requires knowledge of the physicochemical properties and structure of petroleum thin films, including the formation of 3D aggregates.4-6 Aggregation, emulsification, and deposition of petroleum fractions cause severe difficulties in petroleum processing7 and in recovering it from an oil spill in the sea.8 One of the most suitable methods to investigate films at the air/ water interface is the Langmuir film technique, which allows research of thin films deposited onto aqueous subphases. The *To whom correspondence should be addressed. E-mail: divinomar@ gmail.com (1) Nicodem, E. D.; Guedes, L. B. C.; Fernandes, C. Z. M.; Severino, D.; Correa, J. R.; Coutinho, C. M.; Silva, J. Prog. React. Kinet. Mech. 2001, 26, 219– 238. (2) Jacquot, F.; Guiliano, M.; Doumenq, P.; Munoz, D.; Mille, G. Chemosphere 1996, 33, 671–681. (3) Nicodem, E. D.; Guedes, L. B. C.; Fernandes, C. Z. M.; Correa, J. R. Biogeochemistry 1997, 39, 121–138. (4) Dalmazzone, C.; Noik, C.; Komunjer, L. S.P.E. J. 2005, 44–53. (5) Garrett, W. D. J. Rech. Atmos. 1974, 8, 555–562. (6) Van Nierop, A. E.; Ajdari, A.; Stone, A. H. Phys. Fluids 2006, 18, 381051– 381054. (7) Rogel, E. Langmuir 2004, 20, 1003–1012. (8) Walker, H. A.; Kucklick, H. J.; Michel, J. Pure Appl. Chem. 1999, 71, 67–81. (9) Davies, J. T.; Rideal, E. K. Interfacial Phenomena; Academic Press: New York, 1961.
Langmuir 2009, 25(21), 12585–12590
Langmuir films are commonly prepared from amphiphilic compounds or mixed with specific substances,9,10 with the molecules organizing themselves in two or three dimensions in a similar fashion to the gaseous, liquid, and solid states of matter, in order to minimize the interfacial free energy.11 In addition to amphiphilic molecules with well-defined polar and nonpolar groups, polymers and other complex molecules may also form organized thin films.12,13 A great deal of information can be obtained about intermolecular or interparticle interactions even for these unconventional samples, despite the difficulty in analyzing the isotherms of films made with complex molecules because of the tendency of aggregation and formation of 3D structures.14 For example, the packing of polymer films on the water surface is determined primarily by cohesive forces of the hydrophobic chains with water, frequently favoring 3D conformations.15 Also, during compression, monolayers of various polymers sustained long-lasting pressure gradients across the film, producing inhomogeneities.13,15 Several components of petroleum are amphiphilic molecules with complex chemical structures, such as resins and asphaltenes, which contribute to petroleum spreading spontaneously on the water surface.4,16,17 The properties of these components in films18 (10) Ferreira, M.; Caetano, W.; Itri, R.; Tabak, M.; Oliveira, N. O., Jr. Quim. Nova 2005, 28, 502–510. (11) Harkins, D. W. J. Chem. Phys. 1941, 9, 552–568. (12) Zhang, Y. L.; Lawrence, S.; Xu, Z.; Masliyah, H. J. J. Colloid Interface Sci. 2003, 264, 128–140. (13) Dhanabalan, A.; Balogh, D. T.; Riul, A., Jr.; Giacometti, J. A.; Oliveira, O. N., Jr. Thin Solid Films 1998, 323, 257–264. (14) Zhang, Y. L.; Xu, Z.; Masliyah, H. J. Langmuir 2003, 19, 9730–9741. (15) Oliveira, O. N., Jr.; Raposo, M.; Dhanabalan, A. Handb. Surf. Interfaces Mater. 2001, 4, 1–63. (16) Leblanc, R. M.; Thyrion, F. C. Fuel 1989, 68, 260–262. (17) Singh, B. P. Energy Sources 1997, 29, 443–447. (18) Cadena-Nava, R. D.; Cosultchi, A.; Ruiz-Garcia, J. Energy Fuels 2007, 21, 2129–2137.
Published on Web 08/24/2009
DOI: 10.1021/la9017757
12585
Article
and solutions19 have been widely discussed. Petroleum is normally divided into fractions upon dissolving in pentane or heptane (classified as maltenes and asphaltenes) as well as in aliphatic, aromatic, and polar resins. Petroleum without asphaltene is called maltene.20,21 Asphaltenes are known to form aggregates even in organic solvents, whereas the resins stabilize asphaltene emulsions due to their amphiphilic, surface-active properties.18,22-25 Studies on the structure and properties of Langmuir films made of specific fractions of petroleum, such as asphaltenes and resins, can be found in the literature.12,19,25 Asphaltenes and resins are the main surface-active molecules that lead to undesired emulsions during oil exploitation. Early in the 1990s, Singh and coworkers investigated the effect of subphase properties and emulsifier additives on the surface properties of purified fractions of petroleum.26 Lobato and coauthors observed relatively rigid thin films of asphaltenes with large hysteresis in surface-pressure curves measured at the oil/water interface.19 The addition of additives reduced rigidity, in the same way as resins do, which seems to be important for breaking emulsions.27 The demulsifiers’ properties in petroleum thin films have also been studied.14,17 Aggregation has been proven in separate studies to govern the properties of Langmuir films of asphaltene and additives.7,28-30 The mechanism of self-assembling seems to be a stepwise process instead of a phase transition at a critical concentration,12,19,31 but this is not totally clear yet. Although the properties of thin films made of specific fractions of petroleum are understood to some extent, as mentioned above, the same is not true for films of petroleum samples. In this study, we have analyzed surface pressure-area isotherms and the threshold of aggregation and domain formation as a function of surface pressure, in order to obtain a comprehensive understanding of the surface properties of a Brazilian petroleum sample. Because inhomogeneities exist in the Langmuir films, it is essential to analyze the film morphology, which was done here with Brewster angle microscopy (BAM)32-36 and fluorescence microscopy.37 The latter is convenient because petroleum’s fluorescence is a useful property to find oil spread on the water.38-40 Almost all petroleum fractions have some fluorescence, whose (19) Lobato, M. D.; Pedrosa, J. M.; Hortal, A. R.; Martinez-Haya, B.; Lebron-Aguilar, R.; Lago, S. Colloids Surf., A 2007, 298, 72–79. (20) Speight, J. G. Chem. Technol. Pet. 1991, 44, 760. (21) Solovyev, A.; Zhang, L.; Xu, Z.; Masliyah, J. H. Energy Fuels 2006, 20, 1572–1578. (22) McLean, D. J.; Kilpatrick, K. P. J. Colloid Interface Sci. 1997, 189, 242– 253. (23) M-H.; Yang, X.; Sj€oblom, J. Colloid Polym. Sci. 1998, 276, 800–809. (24) Ghosh, K. A.; Srivastava, K. S.; Bagchi, S. Fuel 2007, 86, 2528–2534. (25) Ese, M.-H.; Galet, L.; Clausse, D.; Sj€oblom, J. J. Colloid Interface Sci. 1999, 220, 293–301. (26) Singh, B. P.; Pandey, P. B. Indian J. Technol. 1991, 19, 783–788. (27) Rocha Junior, L. C.; Ferreira, M. S.; Ramos, A. C. S. J. Pet. Sci. Eng. 2006, 51, 26–36. (28) Ramos, A. C. S.; Haraguchi, L.; Notrispe, F. R.; Loh, W.; Mohamed, S. R. J. Pet. Sci. Eng. 2001, 32, 201–216. (29) Murgich, J.; Abanero, J. A. Energy Fuels 1999, 13, 278–286. (30) Long, J.; Xu, Z.; Masliyah, J. H. Langmuir 2007, 23, 6182–6190. (31) Albuquerque, F. C.; Nicodem, D. E.; Krishnaswamy, R. Appl. Spectrosc. 2003, 57, 805–810. (32) Henon, S.; Meunier, J. Rev. Sci. Instrum. 1991, 62, 936–939. (33) Honig, D.; Mobius, D. J. Phys. Chem. 1991, 95, 4590–4592. (34) Brandal, O.; Vioitala, T.; Sjoblom, J. J. Dispersion Sci. Technol. 2007, 28, 95–106. (35) Lobato, D. M.; Pedrosa, M. J.; Mobius, D.; Lago, S. Langmuir 2009, 25, 1377–1384. (36) Nordgard, L. E.; Landsem, E.; Sjoblom, J. Langmuir 2008, 24, 8742–8751. (37) McConnell, H. M. Annu. Rev. Phys. Chem. 1991, 42, 171–195. (38) Burlamacchi, P.; Cecchi, G.; Mazzinghi, P.; Pantani, L. Appl. Opt. 1983, 22, 48–53. (39) Hengstermann, T.; Reuter, R. Appl. Opt. 1990, 29, 3218–3227. (40) Karpicz, R.; Dementjev, A.; Kuprionis, Z.; Pakalnis, S.; Westphal, R.; Reuter, R.; Gulbinas, V. Appl. Opt. 2006, 45, 6620–6625.
12586 DOI: 10.1021/la9017757
Vieira et al.
intensity depends on the excitation wavelength, concentration, and observation mode31,41-43 due to preabsorption, reabsorption, and inner filter effects. These kinds of effects are not expected in petroleum films on the water surface unless there is self-aggregation at the molecular level.
Materials and Methods Medium API grade Brazilian petroleum from Bacia de Campos, Rio de Janeiro, Brazil, was kindly donated by Prof. David E. Nicodem from Universidade Federal do Rio de Janeiro. This petroleum sample contains 5% asphaltenes, 38% aromatic compounds, 14% polar resins, 43% aliphatics, and 0.5% sulfur compounds. Its viscosity is 65.2 cSt (at 30 °C), with a density of 0.8595 g 3 mL-1. Chloroform (99.8%) was acquired from Labsynth. Acridine orange (85% pure, from Sigma) was dissolved in water at 3.63 mM, and this stock solution was used to produce a fluorescent subphase. Petroleum samples were spread on water in a KSV-5000 Langmuir trough (530 (l) 150 (w) mm2) located in a clean room kept at 21.5 °C. Ultrapure water with a resistivity of 18.2 MΩ 3 cm, pH ∼ 6, and a surface tension of 72.8 mN/m at 25 °C was supplied by a Milli-RO coupled to a Milli-Q purification system from Millipore. The surface pressure and surface potential were measured with a Wilhelmy plate and a Kelvin probe, respectively. Films were obtained by spreading chloroform solutions of petroleum at concentrations of 2.2 and 8.7 mg/mL at the air/water interface with a microsyringe. Film compression was carried out after complete solvent evaporation (∼15 min) at a speed of 25 mm/min. Because of the complex constitution of petroleum molecules, with different sizes, shapes, and no average molar mass available, the isotherms were plotted with pressure or potential versus mass of petroleum spread (cm2/mg). This procedure is commonly used in the analysis of asphaltene films spread on the water surface.19 The hysteresis curves were obtained after preparation of the monolayer by spreading 500 μL of a chloroform solution (2.2 mg/mL). Three cycles of compression/decompression with a barrier speed of 25 mm/min were performed. Fluorescence microscopy experiments were carried out in a homemade Teflon trough (114 (l) 39 (w) mm2) mounted in the stage of an Olympus B-Max epifluorescence microscope. The movable barrier was driven by a motor at a speed of 0.78 mm/min. Aliquots of 25 μL of petroleum dissolved in chloroform at 2.2 mg/ mL were spread on the water subphase. The excitation source was a low-pressure Hg lamp (100 W), with a band-pass dichroic filter (400-440 nm). Petroleum film with Acridine in the subphase was excited using a proper filter (520-550 nm), and the emission was registered above 580 nm. The image acquisition system allowed detection of emitted light above 475 nm. The liquid interface was imaged using a 20 objective, with a Hamamatsu-SIT camera, 1 mLUX. The images were recorded continuously, taken concomitantly with the surface pressure measurements. Monolayer compression was performed after complete solvent evaporation (∼15 min). The fluorescence images, captured with a K-7 film recorder, were extracted with Power DVD software and treated on the Image J (from USA NIH, public domain) software. This procedure was also used in the experiment for identifying the presence of water domains, but with Acridine orange in the subphase. (41) Mullins, C. O.; Mitra-Kirtley, S.; Zhu, Y. Appl. Spectrosc. 1992, 46, 1405– 1411. (42) Wang, X.; Mullins, C. O. Appl. Spectrosc. 1994, 48, 977–984. (43) Nicodem, E. D.; Cunha, F. V. M.; Guedes, L. B. C. Appl. Spectrosc. 2000, 54, 1409–1411.
Langmuir 2009, 25(21), 12585–12590
Vieira et al.
Article
Brewster angle microscopy (BAM) experiments were performed with a Nima Langmuir trough in a class 10 000 clean room. The Langmuir film was compressed at a barrier speed of 10 mm/min. The spreading solution for the Langmuir film experiments was obtained by dissolving petroleum in chloroform with a typical concentration of 2.2 and 10.6 mg/mL. Ultrapure water was supplied by a Milli-RO coupled to a Milli-Q purification system from Millipore (resistivity 18.2 MΩ 3 cm). Brewster angle microscopy measurements were performed with a BAM2 apparatus from Nanofilm Technology, which contains a digital camera to register the images directly from the petroleum film on the water surface. The brightness in the images does not reflect the real thickness of the film because several adjustments had to be made for obtaining the images as they became too bright with the film compression.
Results and Discussion The Langmuir films (we shall use the term Langmuir film rather than Langmuir monolayer because the films formed at the air-water interface may not be a true monomolecular layer) from petroleum exhibit several states upon compression, which depend on the volume of material spread, because the area available on the trough used is not sufficient to cover the whole isotherms. Figure 1 shows an amalgamated plot for the surface pressure and surface potential isotherms obtained in different experiments with varying volumes and concentrations. In subsidiary experiments, we verified that decreasing the barrier speed for compressing and expanding the films had negligible effects on the isotherms. Despite the dispersion in part of the data, only to be expected for Langmuir films of complex molecules, it is seen that condensed films may be formed, with high collapse pressures. In subsidiary experiments, we observed collapse pressures above 50 mN/m (see the Supporting Information). The motivation for showing isotherms from different experiments in the same graph was to highlight the almost negligible influence from the concentration and volume spread, which means that the system is relatively well-behaved in terms of aggregation-as will be further discussed-despite the varied composition of the petroleum sample. The in-plane elasticity (Cs-1) of the film was calculated from the surface pressure curve using the following expression:9,44 Cs-1 ¼ -A
Dπ DA
T
where A is the area and π is the surface pressure. The values of Cs-1 indicated typical liquid-expanded (LE) films with 32.2 mN/m for 200 < area < 550 cm2/mg and liquid-condensed phases (LC) with 193 mN/m for 24 < area < 45 cm2/mg, which are marked in the surface pressure isotherms of Figure 1. The gaseous phase (G) is well-characterized on the plot in a region with negligible surface potential and pressure. The surface potential isotherm in Figure 1 displays a lift-off at larger areas than for the surface pressure, as is common to most Langmuir films.44 This means that the coming together of domains in the Langmuir film occurs upon compression before the rise in surface pressure, and this is consistent with the fluorescence and BAM microscopies study to be reported below. That the surface potential is zero at very large areas means that aggregation-known to occur for petroleum samples, as we shall discuss-is not sufficient to yield nonzero potentials at such areas. (44) Dynarowicz-Latka, P.; Cavalli, A.; Silva Filho, D. A.; Milart, P.; Santos, M. C.; Oliveira, O. N., Jr. Chem. Phys. Lett. 2001, 337, 11–17.
Langmuir 2009, 25(21), 12585–12590
Figure 1. Pressure-area and surface potential-area isotherms for a petroleum Langmuir film prepared in a trough with an area of 792 cm2. The figure combines data from three experiments, with the concentrations (c) and volume spread (Vsp) given in the inset.
Figure 2. Cycles of compression and decompression of a thin film of petroleum on the water surface. A petroleum/chloroform (500 μL, 2.2 mg/mL) solution was spread, for a subphase temperature of 21.5 °C, which allows observing both the gas (G) and the liquid-expanded (LE) states of the film.
Also similar to other Langmuir films, the slope in the surface potential increase is lower when the liquid-expanded state is reached, as shown in Figure 1. Another feature to be noted is the high value of the maximum surface potential, ca. 0.8 V, which indicates the presence of strong polar groups in the petroleum sample, as the surface potential is proportional to the vertical component of the molecular dipole moments. Because neither the surface groups nor their orientation are known precisely, the surface potential data cannot be treated quantitatively. They serve, nevertheless, to show the lack of sharp phase transitions in the Langmuir film. Langmuir films with expanded isotherms, such as those of Figure 1, normally exhibit large hysteresis in the compression-decompression cycles. For macromolecules, for instance, such hysteresis is due to the difficulties in disrupting the aggregates formed during compression, which causes the pressure in the decompression procedure to be much lower than during compression. This expectation was not fulfilled for the petroleum film, as indicated in Figure 2 that shows almost no hysteresis. Therefore, the petroleum sample behaves in a completely different way to that of asphaltenes, for which a large hysteresis owing to irreversible aggregation is well-documented.14,18,19,30 Even more interesting is the shift toward larger areas and increase in the maximum pressure observed in consecutive compression-decompression cycles. In most Langmuir films, the isotherm shift occurs toward lower areas due to loss of material to the subphase and rearrangement. It seems that the packing of the petroleum molecules is such that the whole system becomes DOI: 10.1021/la9017757
12587
Article
Vieira et al.
Figure 3. Fluorescence microscopy images of a petroleum film on the water surface without compressing at 25 °C at the G phase. Images a-e represent typical data acquired during 10 min (pressure at 0 mN/m). The images were treated with Image J software, inverted, and converted into 8 bits, in tones of light and dark and with adjustment of brightness and contrast. Black regions represent highly fluorescent regions of the film.
Figure 4. Fluorescence microscopy images of a petroleum film on the water surface being compressed from 900 cm2/mg (π = 0 mN/m) to 550 cm2/mg (π = 2.6 mN/m) at 25 °C. Images a and b represent the gaseous phase (G), whereas those in images c-e were obtained at the first stages from the liquid-expanded (LE) phase. The same image treatment described in Figure 3 was applied.
more surface-active after two compression-decompression cycles. As it will become clear with fluorescence and BAM images, we expect this to be related to the uptake of water during the compression cycle. A similar shift to larger areas was observed by Dı´ az for asphaltenes, bitumen, and maltenes, but for these, there was strong hysteresis.45 Taken together, the results presented here point to a more stable Langmuir film for the petroleum sample in comparison with the films from purified fractions.18,19,45 This is consistent with the finding that resins and other petroleum components stabilize asphaltenes, by decreasing aggregation, especially the irreversible aggregates.17,22,25 Taking advantage of the natural fluorescence of several petroleum constituents, we were able to characterize, in detail, the domain formation process upon compression of Langmuir films. The images shown in Figures 3-5 are displayed with inversion of contrast. Hence, black regions represent phases rich in emitting materials with excitation at 400-440 nm and emission above 500 nm, whereas light gray regions correspond to phases rich in weakly emitting materials. The choice of this wavelength range for excitation was based on the literature,1,3,31,41-43 in order to obtain the highest intensities for the fluorophores in petroleum. In the G state, at about 900 cm2/mg, as it is usual for this state with a high degree of translational motion, nonsystematic changes are observed. Indeed, Figure 3 displays an inhomogeneous pale-fluor(45) Dı´ az, E. M.; Montes, J. F.; Galan, A. M. Energy Fuels 2007, 21, 3455–3461.
12588 DOI: 10.1021/la9017757
escent field and a few small fluorescent domains. In addition, in some cases, larger, movable, and highly fluorescent domains are observed. Because of the translation motion and collisions, emitting molecules may form reversible separate phases rich in emitting material, which appear as black spots ranging from 20 to 80 μm. Hereafter, these spots will be referred to as large microdomains (LMDs). The process of appearance and disappearance of these fluorescent domains is dynamic, but the overall features of Figure 3 were seen to be preserved upon filming the surface for ∼10 min. An organization in the nanometer scale has been related to the formation of a series of asphaltene structures (2-100 nm), including micelle-like, disklike, and ribbons, observed with atomic force microscopy in LB films of asphaltene transferred onto mica.18 The optical technique used here, viz. fluorescence microscopy, cannot resolve nanometer-sized structures, but it is likely that the inhomogeneous background of Figure 3 is made of small domains in this size range, which we shall refer to as submicrodomains (SMDs). That the domain formation increases fluorescence may be explained as follows. The increase in fluorescence in a 2D system depends on the surface density of emitting molecules and on their fluorescence quantum yields.1,31,46 Intermolecular interactions among emitting molecules may cause either an (46) Hiranita, T.; Nakamura, S.; Kawachi, M.; Courrier, H. M.; Vandamme, T. F.; Kraft, M. P.; Shibata, O. J. Colloid Interface Sci. 2003, 265, 83–92.
Langmuir 2009, 25(21), 12585–12590
Vieira et al.
Article
Figure 5. Fluorescence microscopy images of a petroleum film on the water surface obtained when the barriers were stopped at the minimum area/mg (200 cm2/mg, π = 8.0 mN/m) at 25 °C, at the LE phase. Images a-e represent the more advanced stages of the liquid-expanded phase (LE). The same image treatment of Figure 3 was applied.
increase or a decrease in emission quantum yields,31,47 depending on the proximity and conformation of the molecules. Quenching can occur in the ground state at the molecular level48 and after excitation by diffusion collision that can be chemical or physical quenching.47 Some types of J aggregates show increased fluorescence due to exciton coupling.49 The increased fluorescence observed here points to complexation that may be stabilized by hydrogen bonds between molecules instead of π stacking. Our results indicate that the formation of domains led to increased fluorescence, in relation to the background, owing to an increase in the surface density of emitting molecules. We suppose that the fluorescent species are dispersed within the domains, which have microscopic dimensions, being kept apart by polar resins, asphaltenes, and other petroleum components that form the domains. Changes in the film structure during compression from the G phase to the G-LE transition are shown in Figure 4, featuring, in the latter, a larger number of LMDs of different sizes and fluorescence intensities, dispersed in a continuous inhomogeneous field. As the pressure was further increased and the barriers stopped in the middle of the LE phase (200 cm2/mg, π = 8.0 mN/ m), coalescence of SMDs took place, forming an inhomogeneous background of increased fluorescence and large white spots, as seen in Figure 5. The latter spots concentrate in the contour of black domains, which seem to enable redispersion of the white domains. The image in Figure 5 is interpreted as follows: upon lateral compression and consequent increase of surface density, a phase separation of low-emitting material occurs. The emitting material then concentrates on the contours of the domains, thus increasing the line perimeter of mesoscopic nonemitting domains, analogous to the formation of a bidimensional emulsion. The low-fluorescent domains, represented by white areas on the images of Figure 5, correspond to nonfluorescent molecules at this range of excitation/emission of the spectrum, possibly with some emulsified water. The low-emitting LMDs observed in Figures 3-5 are only a little more fluorescent than the G phase (see the Supporting Information), which indicates a slightly larger concentration of (47) Wayne, C. E.; Wayne, R. P. Photochemistry; Oxford University Press: Oxford, UK, 1996. (48) Severino, D.; Junqueira, H. C.; Gugliotti, M.; Gabrielli, D. S.; Baptista, M. S. Photochem. Photobiol. 2003, 77, 459–468. (49) Guralchuk, G. Y.; Katrunov, I. K.; Grynyov, R. S.; Sorokin, A. V.; Yefimova, S. L.; Borovoy, I. A.; Malyukin, Y. V. J. Phys. Chem. C 2008, 112, 14762–14768.
Langmuir 2009, 25(21), 12585–12590
fluorescent molecules upon compression. A hypothesis to explain this low-emitter LMD, which would be in accordance with the small hysteresis in Figure 2, is that some water is incorporated and stabilized by the amphiphilic compounds. To check this hypothesis, we added a highly fluorescent hydrophilic probe (Acridine orange) into the aqueous subphase. We observed fluorescence from Acridine orange in these domains, indicating that, indeed, the water solution was stabilized in the petroleum film at high surface pressures (data not shown). Furthermore, the BAM images in Figure 6 also display dark areas for the LE and LC phases, which means that there are no organic molecules in those regions. If organic compounds were at the interface, the refractive index would have changed, thus causing light to reach the camera and an image of the interface structure would be formed. This is because a large optical contrast between pure water (black regions) and the spots covered with film molecules (white regions) can be achieved.32,33 Overall, the BAM images confirm the presence of distinct phases upon compression inferred from the fluorescence microscopy data. Therefore, one may conclude that water is incorporated in the petroleum film, which led to an increase in the maximum pressures for consecutive compression-decompression cycles in Figure 2. Furthermore, such water incorporation may have an important bearing on the emulsion formation. The observations in Figures 3-5 were better quantified by measuring the emitted light intensity/μm2 and by counting the average number of LMDs and corresponding fluorescence intensities. For more advanced stages of compression, both the intensity of the fluorescent background and the number of LMDs increased, respectively, for the G-LE and LE phases (Supporting Information). These increases were probably caused by coalescence of SMDs to form LMDs, when the film reached the LE-LC boundary states. Interestingly, instead of a straightforward increase in domain formation, the LE phase was characterized by phase separation and coalescence of different components, explaining the change in surface potential behavior in Figure 1. The changes upon compression were reversible; that is, upon increasing the area from LE to G, similar features appeared as going from G to LE. This reversibility is consistent with the small hysteresis in Figure 2 for the surface pressure isotherms and differs from the irreversible behavior of asphaltene films.18 Again, the reversibility supports the hypothesis of greater stabilization of the petroleum film compared with films of its fraction. The results shown here, particularly the images, allow us to infer that this petroleum sample forms an inhomogeneous DOI: 10.1021/la9017757
12589
Article
Vieira et al.
Figure 6. BAM images of the petroleum film at the gas (G), liquid-expanded (LE), and liquid-condensed (LC) phases of the surface pressure isotherm. * represents surface pressure around 20 mN/m. ** represents surface pressure around 40 mN/m.
structure at the air/water interface, with domains appearing at all compression stages. Increasing the pressure in the G phase leads to a stepwise increase in film organization due to formation of larger, more numerous aggregated domains. In the G phase, a few highly fluorescent domains appear. The formation and disappearance of these domains is a dynamic process indicating the larger fluidity of the petroleum film samples compared with films made of single components. The domains appear to be more surface-active, which makes the domain formation a reversible process and allows stabilization of water in the film. In the LC phase, coalescence and phase separation is the main feature leading to higher collapse pressures.
Conclusions Samples of petroleum were spread onto the water surface, thus forming highly stable Langmuir films, whose surface pressure isotherms displayed typical phases of a monolayer. Indeed, the in-plane elasticity for the various stages of compression indicated gas, liquid-expanded, and liquid-condensed phases. Important features of the petroleum Langmuir films were the lack of hysteresis in compression-decompression cycles and the high collapse pressures. Both results demonstrate that these films are more stable than those of purified fractions of petroleum, such as asphaltenes. The compression of the film in the Langmuir trough led to domain formation that could be studied with
12590 DOI: 10.1021/la9017757
fluorescence microscopy. An increasing number of fluorescent domains with enhanced fluorescence indicated that the domains are formed by fluorescent chromophores, which absorb in the blue region and emit in the green. Furthermore, the domain formation was reversible, once again confirming that petroleum films behave differently from those made with asphaltenes investigated in the literature. It is also concluded that, to probe the effect of additives in the formation of emulsions or spreading of petroleum films, it is equally important to perform studies with whole petroleum samples. The visualization of water stabilization in petroleum films by surface fluorescence microscopy and BAM could be a new tool to study emulsion formation in petroleum samples. Acknowledgment. This work was supported by FAPESP, CNPq, and CAPES (Brazil). D.S. thanks Farma Service BioExtract for a postdoctoral fellowship. Supporting Information Available: Pressure-area isotherm for a petroleum film, graph of intensity per unit area for various states of the Langmuir film, graph of average fluorescent intensities and number of LMDs in each phase during compression of the petroleum film on the water surface, and emission spectra from the petroleum film on the water surface. This material is available free of charge via the Internet at http://pubs.acs.org.
Langmuir 2009, 25(21), 12585–12590
