- Email: [email protected]
A novel technique for improving interferometric determination of emulsion film thickness by digital filtration
Journal of Colloid and Interface Science 306 (2007) 449–453 www.elsevier.com/locate/jcis
Priority Communication
A novel technique for improving interferometric determination of emulsion film thickness by digital filtration Stoyan I. Karakashev a , Anh V. Nguyen a,∗ , Emil D. Manev b a Discipline of Chemical Engineering, School of Engineering, The University of Newcastle, Callaghan, New South Wales 2308, Australia b Department of Physical Chemistry, Faculty of Chemistry/Sofia University, 1J. Bourchier Ave, 1164 Sofia, Bulgaria
31 August 2006; accepted 11 November 2006 Available online 21 November 2006
Abstract Determination of the thickness of emulsion films by using the film interferometric images is usually less accurate than that of foam films, due to the close values of the refractive indices of the liquid film and adjacent liquid phases (hence, low contrast and high level of noise at high magnification). A new technique was developed to improve the thickness determination by obtaining the interferometric images without directly filtering the illuminating light, as is usually done in the classical Scheludko interferometric technique. The new method then uses digital filtration during the off-line image post-processing to obtain monochromatic interferometric images required for the thickness determination. The technique was tested with foam films stabilised by sodium dodecyl sulfate and successfully applied to determine thickness of toluene–water– toluene emulsion films using the green and red digital filters. Results for emulsion film thickness determined by either the green or red digital filtration are comparable, thus validating the new technique developed here for emulsion films. © 2006 Elsevier Inc. All rights reserved. Keywords: Emulsion film; Foam film; Interferometry
1. Introduction For many years, microinterferometry has successfully been used to determine thickness of liquid thin films between a solid surface and an air bubble (wetting films), or between fluid– fluid interfaces (foam films between two bubbles or emulsion films between two droplets) [1–11]. Generally, a monochromatic (usually green) light of known wavelength is shined on and reflected from the film surfaces, consequently undergoing interference and producing a series of sequential dark and light stripes (interferometric extrema), known as the Newton rings. The resulting interferometric pattern is registered and used to convert the photocurrent intensity of the monochromatic light with the known wavelength into thickness. The amplitude of the photocurrent intensity, that is the difference in its maximum and minimum values, is critical to the thickness calculation: the larger the difference, the higher the accuracy of the film thick* Corresponding author. Fax: +612 492 16920.
E-mail address: [email protected] (A.V. Nguyen). 0021-9797/$ – see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2006.11.023
ness calculated. The intensity amplitude is determined by the contrast of the interferometric images. In the traditional microinterferometric measurements using the Scheludko cell [2,3], monochromatic light is focused onto a small portion of the film area. The photocurrent signal obtained is amplified by a photomultiplier before recording. In the process of film thinning, the decrease in the film thickness on these spots of the film area is recorded in the form of interferograms. The temporal variation of the photocurrent intensity in the interferograms is used to calculate the film thickness as a function of time. The kinetic description of the film thinning thus obtained is more precise for very small films (with radius smaller than 0.1 mm), which are usually planar at the film thickness smaller than 100 nm [12]. The larger films exhibit significant corrugations on the film surfaces and dimples [13]. Therefore, the film thickness in such cases depends on the location over the film surface, on which the monochromatic light beam is focused. In the recent years, the interferometric measurements underwent new developments. Instead of being focused on a small area on the film surface, the beam of monochromatic light
450
S.I. Karakashev et al. / Journal of Colloid and Interface Science 306 (2007) 449–453
can be focused on the whole film surface. The interferometric images thus obtained are captured by high-speed video CCD microscopy and transferred to a computer for recording and off-line analysis. In some cases, it can be done with a linescan camera [14] capable of scanning a chosen line (with a thickness ca. 0.5 µm) throughout the film, thus producing a series of interferograms in the suitable digital format. They correspond to the variation of the film thickness (along the scanned line) versus time. Otherwise, a high-speed video CCD camera can be used to capture the whole film, producing a series of sequential interferometric images, which are transferred to a computer and stored in the analogous format. The images are then processed using image analysis software and converted into the digital format for calculating the film thickness. This modification of the interferometric methodology can increase the accuracy of the thickness calculation significantly (in some cases up to 40 times). Therefore, the change in the transient film thickness can be obtained with higher accuracy. Other advances of this technique include the digital processing of the analogous interferometric images by means of the sophisticated development in image analysis software in recent years. The interferometric method of Scheludko described above has been applied to study the emulsion films in many works [15–24]. A common problem of these investigations, in particular, when the interferometric images are obtained using a CCD camera, is the low contrast of the interferometric images due to the similar values of the refractive indices of the film and adjacent phases which results in small difference between the interferometric minimum and maximum, and consequently reduces the accuracy of the thickness determination. The aim of this paper is to present a new technique for improving the interferometric determination of emulsion film thickness. Unlike the common Scheludko interferometric method, the new technique does not use optical filtering to produce the monochromatic interferograms or images. Instead, a CCD high-speed camera is used to record the interferometric images of the whole transient films without filtering the light. The interferometric images are then filtered in the off-line postprocessing stage using the digital filtration. The monochromatic interferograms can be obtained with different digital filters with different wavelengths available in image analysis software. 2. Experimental The experimental setup is based on the classical Scheludko microinterferometric cell and schematically shown in Fig. 1. The major units of the setup includes (Figs. 1 and 2): • A Scheludko cell [3] for producing the thin liquid films. • An inverted microscope (Epihot 200, Nikon, Japan) with the light source for illuminating and observing the interferometric images of the film in the reflected light. • A high-speed video camera (Phantom 4, Photo-Sonics Inc., USA) for capturing the interferometric images. • A computer for controlling the high-speed video camera and recording the interferometric images.
Fig. 1. Schematic of the experimental microinterferometric setup with a Scheludko cell, a metallurgical inverted microscope, and a high-speed video camera system.
Emulsion films were produced by immersing a droplet of water, formed inside the film holder of the Scheludko cell, into the toluene phase contained in the lower part of the cell. In the case of foam films, a droplet of the aqueous solution of surfactant was used in place of the water drop and the bottom of the cell was kept empty. A capillary tube was used to connect the film holder with a gastight microsyringe for controlling the amount of water inside the droplet. The microscopic film was formed between the surfaces of the double concave meniscus by pumping out the liquid from the droplet as shown in Fig. 1 and was initiated in this manner to drain. The radius of the film was dependent on the amount of liquid withdrawn from the droplet. A removable monochromatic green filter (with wavelength λ = 546 nm) is commonly placed on the light path prior to the camera to produce monochromatic interferograms. Such the interferograms are used to calculate film thickness of foam films. In the case of toluene–water–toluene emulsion films, the images of the thinning films were totally impeded by the filter. Consequently, the filter was removed to obtain (unfiltered) interferometric images. These images were processed off-line with a digital monochromatic filter using image analysis software Optimas (Version 6.5, Optimas Inc., USA) to produce monochromatic interferograms (fringes) for calculating the film thickness. Narrow strips passing through the centres of the monochromatic interference fringes were selected and digitalised to produce the photocurrent versus radial distance. The photocurrent intensity could be increased by selecting wider strips. The other advantages include the use of different digital filters with different wavelength and the possibility of averaging the photocurrent profiles obtained in different directions, giving an averaged profile representing the (planar) film. The digitisation process used to obtain single photocurrent profiles was automatically carried out using a special macro developed by us in Optimas 6.5. The film thickness profile along the chosen line was calculated using the Scheludko interferometric equation described by Eq. (1), which accounts for the light interference
S.I. Karakashev et al. / Journal of Colloid and Interface Science 306 (2007) 449–453
451
Fig. 2. A photograph showing the interferometric setup described in Fig. 1.
by multiple reflections from the two film surfaces [3,25]:
3. Results and discussion
λ (1 + R)2 h= lπ ± arcsin . 2πn0 (1 − R)2 + 4R
3.1. Validation of the new technique with foam films (1)
In Eq. (1), λ is the wavelength of the monochromatic light and l = 0, 1, 2, . . . is the order of interference. R = (n0 − n1 )2 /(n0 + n1 )2 is the Fresnel reflection coefficient, where n0 and n1 are the refractive indices of the film liquid (water) and the fluid phase (oil or air), respectively. = (I − Imin )/(Imax − Imin ), where I is the instantaneous photocurrent intensity, and Imin and Imax are the minimum and maximum of I . For toluene–water–toluene emulsion films, n1 = 1.496 and n0 = 1.333. For air–water–air foam films, n1 = 1 and n0 = 1.333. The interferometric method is based on the reflectivity of the film surfaces. The monochromatic light being reflected by the two film surfaces undergoes a phase shift which depends on the local film thickness, and leads to the summation (maxima) or subtraction (minima) of the intensities of reflected light. The difference between the highest maximum and the deepest minimum determines the image contrast. The higher the difference, the higher the accuracy of the thickness determination. The reflection by the film surfaces is quantitatively described by the Fresnel reflection coefficient, R. For air–water–air foam films, R is about 0.02 but for toluene–water–toluene emulsion film, R is about 0.0033. One can see that the image contrast of the emulsion films is about 6 times lower than the contrast of foam films. The utilization of the conventional monochromatic green filter in the classical Scheludko microinterferometric method leads to reduction in the light intensity and the contrast of the interferometric images, and the decrease in the accuracy of the thickness calculation discussed above.
It is important to check if the new technique based on the digital filtration of the multiple wavelength interferometric images can provide similar thickness determined by the conventional Scheludko microinterferometry. The comparison was done with foam films, for which both the conventional and new techniques give reasonably good image contrast for determining thickness. Transient drainage of foam films produced from aqueous solution of 1 × 10−5 mol/L sodium dodecyl sulfate (SDS) was investigated by applying both techniques. The conventional monochromatic green filter (λ = 546 nm) was employed in the classical Scheludko technique. The conventional monochromatic green filter was not employed in the new technique but the digital filtration was carried out using two wavelengths: λ = 546 nm (green) and λ = 650 nm (red). Typical monochromatic images of the SDS foam film obtained by the classical Scheludko technique and by the new techniques using green and red digital filtering are shown in Fig. 3. Clearly, the contrast of the left-hand side image obtained by the classical Scheludko technique is significantly lower than the contrast of the middle and right-hand side images obtained by the new techniques using the digital filtration. Fig. 4 shows the transient film thickness for 1 × 10−5 mol/L SDS foam films obtained with the classical Scheludko and new techniques. Two films of the same radius (about 0.065 mm) were chosen. As it can be observed, the transient thicknesses practically coincide, validating the newly developed technique, based on the digital monochromatic filtration. Fig. 5 shows the comparison for the transient film thickness data, determined with the new technique for one and the same foam film of 1 × 10−5 mol/L SDS solution by using two (green and red) digital monochromatic filters. The difference between the two sets
452
S.I. Karakashev et al. / Journal of Colloid and Interface Science 306 (2007) 449–453
Fig. 3. Interferometric image of a 1 × 10−5 mol/L SDS foam film obtained by the classical Scheludko technique using the conventional monochromatic green filter (left) and by the new technique with digital filtration using either the green (middle) or red (right) digital filter.
Fig. 6. Unfiltered (left) and digitally filtered (right) interferometric images of a stable toluene–water–toluene emulsion film obtained by the new technique using a digital green filter. The interferometric images (not shown) obtained by classical method were dark and noisy. Fig. 4. Transient thickness for 1 × 10−5 mol/L SDS foam films obtained with the classical Scheludko and new techniques. The film radius was about 0.065 mm.
Fig. 5. Transient thickness for one 1 × 10−5 mol/L SDS foam film obtained with the new technique using the green and red digital monochromatic filter. The film radius was about 0.077 mm.
of data is smaller than 1%, showing that, as expected, the new technique yields accurate results, independently of the wavelength of the digital filter. 3.2. Application of the new technique to determine emulsion film thickness The fresh emulsion films formed immediately after the immersion of the film holder (containing a water droplet) in the toluene phase were unstable. The stable emulsion films were
obtained after about 15 min of duration. The increase in the film stability was due to the partial solubility of toluene in water. For example, the solubility of the toluene in water at ambient temperature (20–25 ◦ C) is about 0.53 g/L [19]. Therefore, the film subsurface contain structured layers of water and toluene molecules, which are obviously able to damp the surface capillary waves causing the film instability [3]. Interferometric images of the stable toluene–water–toluene emulsion films obtained with CCD camera by the classical Scheludko technique were dark and noisy, hardly showing any interferometric pattern, because of the very low contrast. Transient and stable toluene–water–toluene emulsion films were investigated using the new technique. The unfiltered and filtered interferometric images of a transient toluene–water– toluene emulsion film obtained by the new technique are shown in Fig. 6. The filtered interferometric images shown in the figure were used to digitise and calculate the transient film thickness shown in Fig. 7. Dimple formation was formed in the transient emulsion film. 4. Conclusions Due to the small difference in the refractive indices of the film and adjacent phases, the interferometric images of toluene– water–toluene emulsion films obtained by the classical Scheludko interferometry were not suitable for correct determination of the emulsion film thickness. A new technique was developed to improve the thickness determination by obtaining the interferometric images without physically filtering the light. The video images were then filtered using the digital filtra-
S.I. Karakashev et al. / Journal of Colloid and Interface Science 306 (2007) 449–453
453
References
Fig. 7. Transient thickness profile for a toluene–water–toluene film obtained from the digitally filtered interferometric image shown on the right-hand side of Fig. 6.
tion to obtain monochromatic interferometric images required in the thickness determination. The technique was tested with SDS foam films: The transient thicknesses for SDS foam films, determined by the classical Scheludko interferometry and the new digital filtration technique coincide within the limits of the experimental error. Finally, the new technique was also successfully applied to determine the thickness of toluene–water– toluene emulsion films using the green and red digital filters. The emulsion film thicknesses determined by either the green or red digital filtration coincide again within the limits of the experimental error, thus validating the new technique developed for emulsion films. Acknowledgments The authors gratefully acknowledge the Australian Research Council for financial support through a Discovery grant. Emil Manev is grateful also for the support by the MCRTN-SOCON Project “Self-Organization Under Confinement.”
[1] B.V. Deryagin, M. Kusakov, Bull. Acad. Sci. USSR (Class Sci. Math. Nat. Ser. Chem.) (1937) 1150. [2] A. Scheludko, D. Ekserova, Kolloid Z. 165 (1959) 148. [3] A. Scheludko, Adv. Colloid Interface Sci. 1 (1967) 391. [4] D. Platikanov, J. Phys. Chem. 68 (1964) 3619. [5] H.J. Schulze, Colloid Polym. Sci. 254 (1976) 438. [6] A.D. Barber, S. Hartland, Trans. Inst. Chem. Eng. 53 (1975) 106. [7] R. Tsekov, P. Letocart, H.J. Schulze, Langmuir 16 (2000) 8206. [8] J.E. Coons, P.J. Halley, S.A. McGlashan, T. Tran-Cong, Adv. Colloid Interface Sci. 105 (2003) 3. [9] E.D. Manev, S.V. Sazdanova, D.T. Wasan, J. Dispersion Sci. Technol. 5 (1984) 111. [10] E.D. Manev, A.V. Nguyen, Adv. Colloid Interface Sci. 114–115 (2005) 133. [11] J.K. Angarska, B.S. Dimitrova, K.D. Danov, P.A. Kralchevsky, K.P. Ananthapadmanabhan, A. Lips, Langmuir 20 (2004) 1799. [12] E. Manev, R. Tsekov, B. Radoev, J. Dispersion Sci. Technol. 18 (1997) 769. [13] E.D. Manev, A.V. Nguyen, Int. J. Miner. Process. 77 (2005) 1. [14] S.I. Karakashev, A.V. Nguyen, E.D. Manev, C.M. Phan, Colloids Surf. A Physicochem. Eng. Aspects 262 (2005) 23. [15] E. Manev, M. Buleva, Colloid Polym. Sci. 256 (1978) 375. [16] A.K. Malhotra, D.T. Wasan, AIChE J. 33 (1987) 1533. [17] B.P. Binks, W.G. Cho, P.D.J. Fletcher, Langmuir 13 (1997) 7180. [18] T.D. Dimitrova, F. Leal-Calderon, T.D. Gurkov, B. Campbell, Langmuir 17 (2001) 8069. [19] E.S. Basheva, T.D. Gurkov, N.C. Christov, B. Campbell, Colloids Surf. A Physicochem. Eng. Aspects 282–283 (2006) 99. [20] E. Mileva, B. Radoev, Hydrodynamic interactions and stability of emulsion films, in: D.N. Petsev (Ed.), Emulsions: Structure, Stability and Interactions, Elsevier, Amsterdam, 2004, p. 215. [21] C. Stubenrauch, R. von Klitzing, J. Phys. Condens. Matter 15 (2003). [22] S.D. Taylor, J. Czarnecki, J. Masliyah, J. Colloid Interface Sci. 251 (2002) 149. [23] K.G. Marinova, T.D. Gurkov, T.D. Dimitrova, R.G. Alargova, D.K.W. Smith, Prog. Colloid Polym. Sci. 110 (1998). [24] O.D. Velev, T.D. Gurkov, R.P. Borwankar, J. Colloid Interface Sci. 159 (1993) 497. [25] A.V. Nguyen, H.J. Schulze, Colloidal Science of Flotation, Dekker, New York, 2004.
Priority Communication
A novel technique for improving interferometric determination of emulsion film thickness by digital filtration Stoyan I. Karakashev a , Anh V. Nguyen a,∗ , Emil D. Manev b a Discipline of Chemical Engineering, School of Engineering, The University of Newcastle, Callaghan, New South Wales 2308, Australia b Department of Physical Chemistry, Faculty of Chemistry/Sofia University, 1J. Bourchier Ave, 1164 Sofia, Bulgaria
31 August 2006; accepted 11 November 2006 Available online 21 November 2006
Abstract Determination of the thickness of emulsion films by using the film interferometric images is usually less accurate than that of foam films, due to the close values of the refractive indices of the liquid film and adjacent liquid phases (hence, low contrast and high level of noise at high magnification). A new technique was developed to improve the thickness determination by obtaining the interferometric images without directly filtering the illuminating light, as is usually done in the classical Scheludko interferometric technique. The new method then uses digital filtration during the off-line image post-processing to obtain monochromatic interferometric images required for the thickness determination. The technique was tested with foam films stabilised by sodium dodecyl sulfate and successfully applied to determine thickness of toluene–water– toluene emulsion films using the green and red digital filters. Results for emulsion film thickness determined by either the green or red digital filtration are comparable, thus validating the new technique developed here for emulsion films. © 2006 Elsevier Inc. All rights reserved. Keywords: Emulsion film; Foam film; Interferometry
1. Introduction For many years, microinterferometry has successfully been used to determine thickness of liquid thin films between a solid surface and an air bubble (wetting films), or between fluid– fluid interfaces (foam films between two bubbles or emulsion films between two droplets) [1–11]. Generally, a monochromatic (usually green) light of known wavelength is shined on and reflected from the film surfaces, consequently undergoing interference and producing a series of sequential dark and light stripes (interferometric extrema), known as the Newton rings. The resulting interferometric pattern is registered and used to convert the photocurrent intensity of the monochromatic light with the known wavelength into thickness. The amplitude of the photocurrent intensity, that is the difference in its maximum and minimum values, is critical to the thickness calculation: the larger the difference, the higher the accuracy of the film thick* Corresponding author. Fax: +612 492 16920.
E-mail address: [email protected] (A.V. Nguyen). 0021-9797/$ – see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2006.11.023
ness calculated. The intensity amplitude is determined by the contrast of the interferometric images. In the traditional microinterferometric measurements using the Scheludko cell [2,3], monochromatic light is focused onto a small portion of the film area. The photocurrent signal obtained is amplified by a photomultiplier before recording. In the process of film thinning, the decrease in the film thickness on these spots of the film area is recorded in the form of interferograms. The temporal variation of the photocurrent intensity in the interferograms is used to calculate the film thickness as a function of time. The kinetic description of the film thinning thus obtained is more precise for very small films (with radius smaller than 0.1 mm), which are usually planar at the film thickness smaller than 100 nm [12]. The larger films exhibit significant corrugations on the film surfaces and dimples [13]. Therefore, the film thickness in such cases depends on the location over the film surface, on which the monochromatic light beam is focused. In the recent years, the interferometric measurements underwent new developments. Instead of being focused on a small area on the film surface, the beam of monochromatic light
450
S.I. Karakashev et al. / Journal of Colloid and Interface Science 306 (2007) 449–453
can be focused on the whole film surface. The interferometric images thus obtained are captured by high-speed video CCD microscopy and transferred to a computer for recording and off-line analysis. In some cases, it can be done with a linescan camera [14] capable of scanning a chosen line (with a thickness ca. 0.5 µm) throughout the film, thus producing a series of interferograms in the suitable digital format. They correspond to the variation of the film thickness (along the scanned line) versus time. Otherwise, a high-speed video CCD camera can be used to capture the whole film, producing a series of sequential interferometric images, which are transferred to a computer and stored in the analogous format. The images are then processed using image analysis software and converted into the digital format for calculating the film thickness. This modification of the interferometric methodology can increase the accuracy of the thickness calculation significantly (in some cases up to 40 times). Therefore, the change in the transient film thickness can be obtained with higher accuracy. Other advances of this technique include the digital processing of the analogous interferometric images by means of the sophisticated development in image analysis software in recent years. The interferometric method of Scheludko described above has been applied to study the emulsion films in many works [15–24]. A common problem of these investigations, in particular, when the interferometric images are obtained using a CCD camera, is the low contrast of the interferometric images due to the similar values of the refractive indices of the film and adjacent phases which results in small difference between the interferometric minimum and maximum, and consequently reduces the accuracy of the thickness determination. The aim of this paper is to present a new technique for improving the interferometric determination of emulsion film thickness. Unlike the common Scheludko interferometric method, the new technique does not use optical filtering to produce the monochromatic interferograms or images. Instead, a CCD high-speed camera is used to record the interferometric images of the whole transient films without filtering the light. The interferometric images are then filtered in the off-line postprocessing stage using the digital filtration. The monochromatic interferograms can be obtained with different digital filters with different wavelengths available in image analysis software. 2. Experimental The experimental setup is based on the classical Scheludko microinterferometric cell and schematically shown in Fig. 1. The major units of the setup includes (Figs. 1 and 2): • A Scheludko cell [3] for producing the thin liquid films. • An inverted microscope (Epihot 200, Nikon, Japan) with the light source for illuminating and observing the interferometric images of the film in the reflected light. • A high-speed video camera (Phantom 4, Photo-Sonics Inc., USA) for capturing the interferometric images. • A computer for controlling the high-speed video camera and recording the interferometric images.
Fig. 1. Schematic of the experimental microinterferometric setup with a Scheludko cell, a metallurgical inverted microscope, and a high-speed video camera system.
Emulsion films were produced by immersing a droplet of water, formed inside the film holder of the Scheludko cell, into the toluene phase contained in the lower part of the cell. In the case of foam films, a droplet of the aqueous solution of surfactant was used in place of the water drop and the bottom of the cell was kept empty. A capillary tube was used to connect the film holder with a gastight microsyringe for controlling the amount of water inside the droplet. The microscopic film was formed between the surfaces of the double concave meniscus by pumping out the liquid from the droplet as shown in Fig. 1 and was initiated in this manner to drain. The radius of the film was dependent on the amount of liquid withdrawn from the droplet. A removable monochromatic green filter (with wavelength λ = 546 nm) is commonly placed on the light path prior to the camera to produce monochromatic interferograms. Such the interferograms are used to calculate film thickness of foam films. In the case of toluene–water–toluene emulsion films, the images of the thinning films were totally impeded by the filter. Consequently, the filter was removed to obtain (unfiltered) interferometric images. These images were processed off-line with a digital monochromatic filter using image analysis software Optimas (Version 6.5, Optimas Inc., USA) to produce monochromatic interferograms (fringes) for calculating the film thickness. Narrow strips passing through the centres of the monochromatic interference fringes were selected and digitalised to produce the photocurrent versus radial distance. The photocurrent intensity could be increased by selecting wider strips. The other advantages include the use of different digital filters with different wavelength and the possibility of averaging the photocurrent profiles obtained in different directions, giving an averaged profile representing the (planar) film. The digitisation process used to obtain single photocurrent profiles was automatically carried out using a special macro developed by us in Optimas 6.5. The film thickness profile along the chosen line was calculated using the Scheludko interferometric equation described by Eq. (1), which accounts for the light interference
S.I. Karakashev et al. / Journal of Colloid and Interface Science 306 (2007) 449–453
451
Fig. 2. A photograph showing the interferometric setup described in Fig. 1.
by multiple reflections from the two film surfaces [3,25]:
3. Results and discussion
λ (1 + R)2 h= lπ ± arcsin . 2πn0 (1 − R)2 + 4R
3.1. Validation of the new technique with foam films (1)
In Eq. (1), λ is the wavelength of the monochromatic light and l = 0, 1, 2, . . . is the order of interference. R = (n0 − n1 )2 /(n0 + n1 )2 is the Fresnel reflection coefficient, where n0 and n1 are the refractive indices of the film liquid (water) and the fluid phase (oil or air), respectively. = (I − Imin )/(Imax − Imin ), where I is the instantaneous photocurrent intensity, and Imin and Imax are the minimum and maximum of I . For toluene–water–toluene emulsion films, n1 = 1.496 and n0 = 1.333. For air–water–air foam films, n1 = 1 and n0 = 1.333. The interferometric method is based on the reflectivity of the film surfaces. The monochromatic light being reflected by the two film surfaces undergoes a phase shift which depends on the local film thickness, and leads to the summation (maxima) or subtraction (minima) of the intensities of reflected light. The difference between the highest maximum and the deepest minimum determines the image contrast. The higher the difference, the higher the accuracy of the thickness determination. The reflection by the film surfaces is quantitatively described by the Fresnel reflection coefficient, R. For air–water–air foam films, R is about 0.02 but for toluene–water–toluene emulsion film, R is about 0.0033. One can see that the image contrast of the emulsion films is about 6 times lower than the contrast of foam films. The utilization of the conventional monochromatic green filter in the classical Scheludko microinterferometric method leads to reduction in the light intensity and the contrast of the interferometric images, and the decrease in the accuracy of the thickness calculation discussed above.
It is important to check if the new technique based on the digital filtration of the multiple wavelength interferometric images can provide similar thickness determined by the conventional Scheludko microinterferometry. The comparison was done with foam films, for which both the conventional and new techniques give reasonably good image contrast for determining thickness. Transient drainage of foam films produced from aqueous solution of 1 × 10−5 mol/L sodium dodecyl sulfate (SDS) was investigated by applying both techniques. The conventional monochromatic green filter (λ = 546 nm) was employed in the classical Scheludko technique. The conventional monochromatic green filter was not employed in the new technique but the digital filtration was carried out using two wavelengths: λ = 546 nm (green) and λ = 650 nm (red). Typical monochromatic images of the SDS foam film obtained by the classical Scheludko technique and by the new techniques using green and red digital filtering are shown in Fig. 3. Clearly, the contrast of the left-hand side image obtained by the classical Scheludko technique is significantly lower than the contrast of the middle and right-hand side images obtained by the new techniques using the digital filtration. Fig. 4 shows the transient film thickness for 1 × 10−5 mol/L SDS foam films obtained with the classical Scheludko and new techniques. Two films of the same radius (about 0.065 mm) were chosen. As it can be observed, the transient thicknesses practically coincide, validating the newly developed technique, based on the digital monochromatic filtration. Fig. 5 shows the comparison for the transient film thickness data, determined with the new technique for one and the same foam film of 1 × 10−5 mol/L SDS solution by using two (green and red) digital monochromatic filters. The difference between the two sets
452
S.I. Karakashev et al. / Journal of Colloid and Interface Science 306 (2007) 449–453
Fig. 3. Interferometric image of a 1 × 10−5 mol/L SDS foam film obtained by the classical Scheludko technique using the conventional monochromatic green filter (left) and by the new technique with digital filtration using either the green (middle) or red (right) digital filter.
Fig. 6. Unfiltered (left) and digitally filtered (right) interferometric images of a stable toluene–water–toluene emulsion film obtained by the new technique using a digital green filter. The interferometric images (not shown) obtained by classical method were dark and noisy. Fig. 4. Transient thickness for 1 × 10−5 mol/L SDS foam films obtained with the classical Scheludko and new techniques. The film radius was about 0.065 mm.
Fig. 5. Transient thickness for one 1 × 10−5 mol/L SDS foam film obtained with the new technique using the green and red digital monochromatic filter. The film radius was about 0.077 mm.
of data is smaller than 1%, showing that, as expected, the new technique yields accurate results, independently of the wavelength of the digital filter. 3.2. Application of the new technique to determine emulsion film thickness The fresh emulsion films formed immediately after the immersion of the film holder (containing a water droplet) in the toluene phase were unstable. The stable emulsion films were
obtained after about 15 min of duration. The increase in the film stability was due to the partial solubility of toluene in water. For example, the solubility of the toluene in water at ambient temperature (20–25 ◦ C) is about 0.53 g/L [19]. Therefore, the film subsurface contain structured layers of water and toluene molecules, which are obviously able to damp the surface capillary waves causing the film instability [3]. Interferometric images of the stable toluene–water–toluene emulsion films obtained with CCD camera by the classical Scheludko technique were dark and noisy, hardly showing any interferometric pattern, because of the very low contrast. Transient and stable toluene–water–toluene emulsion films were investigated using the new technique. The unfiltered and filtered interferometric images of a transient toluene–water– toluene emulsion film obtained by the new technique are shown in Fig. 6. The filtered interferometric images shown in the figure were used to digitise and calculate the transient film thickness shown in Fig. 7. Dimple formation was formed in the transient emulsion film. 4. Conclusions Due to the small difference in the refractive indices of the film and adjacent phases, the interferometric images of toluene– water–toluene emulsion films obtained by the classical Scheludko interferometry were not suitable for correct determination of the emulsion film thickness. A new technique was developed to improve the thickness determination by obtaining the interferometric images without physically filtering the light. The video images were then filtered using the digital filtra-
S.I. Karakashev et al. / Journal of Colloid and Interface Science 306 (2007) 449–453
453
References
Fig. 7. Transient thickness profile for a toluene–water–toluene film obtained from the digitally filtered interferometric image shown on the right-hand side of Fig. 6.
tion to obtain monochromatic interferometric images required in the thickness determination. The technique was tested with SDS foam films: The transient thicknesses for SDS foam films, determined by the classical Scheludko interferometry and the new digital filtration technique coincide within the limits of the experimental error. Finally, the new technique was also successfully applied to determine the thickness of toluene–water– toluene emulsion films using the green and red digital filters. The emulsion film thicknesses determined by either the green or red digital filtration coincide again within the limits of the experimental error, thus validating the new technique developed for emulsion films. Acknowledgments The authors gratefully acknowledge the Australian Research Council for financial support through a Discovery grant. Emil Manev is grateful also for the support by the MCRTN-SOCON Project “Self-Organization Under Confinement.”
[1] B.V. Deryagin, M. Kusakov, Bull. Acad. Sci. USSR (Class Sci. Math. Nat. Ser. Chem.) (1937) 1150. [2] A. Scheludko, D. Ekserova, Kolloid Z. 165 (1959) 148. [3] A. Scheludko, Adv. Colloid Interface Sci. 1 (1967) 391. [4] D. Platikanov, J. Phys. Chem. 68 (1964) 3619. [5] H.J. Schulze, Colloid Polym. Sci. 254 (1976) 438. [6] A.D. Barber, S. Hartland, Trans. Inst. Chem. Eng. 53 (1975) 106. [7] R. Tsekov, P. Letocart, H.J. Schulze, Langmuir 16 (2000) 8206. [8] J.E. Coons, P.J. Halley, S.A. McGlashan, T. Tran-Cong, Adv. Colloid Interface Sci. 105 (2003) 3. [9] E.D. Manev, S.V. Sazdanova, D.T. Wasan, J. Dispersion Sci. Technol. 5 (1984) 111. [10] E.D. Manev, A.V. Nguyen, Adv. Colloid Interface Sci. 114–115 (2005) 133. [11] J.K. Angarska, B.S. Dimitrova, K.D. Danov, P.A. Kralchevsky, K.P. Ananthapadmanabhan, A. Lips, Langmuir 20 (2004) 1799. [12] E. Manev, R. Tsekov, B. Radoev, J. Dispersion Sci. Technol. 18 (1997) 769. [13] E.D. Manev, A.V. Nguyen, Int. J. Miner. Process. 77 (2005) 1. [14] S.I. Karakashev, A.V. Nguyen, E.D. Manev, C.M. Phan, Colloids Surf. A Physicochem. Eng. Aspects 262 (2005) 23. [15] E. Manev, M. Buleva, Colloid Polym. Sci. 256 (1978) 375. [16] A.K. Malhotra, D.T. Wasan, AIChE J. 33 (1987) 1533. [17] B.P. Binks, W.G. Cho, P.D.J. Fletcher, Langmuir 13 (1997) 7180. [18] T.D. Dimitrova, F. Leal-Calderon, T.D. Gurkov, B. Campbell, Langmuir 17 (2001) 8069. [19] E.S. Basheva, T.D. Gurkov, N.C. Christov, B. Campbell, Colloids Surf. A Physicochem. Eng. Aspects 282–283 (2006) 99. [20] E. Mileva, B. Radoev, Hydrodynamic interactions and stability of emulsion films, in: D.N. Petsev (Ed.), Emulsions: Structure, Stability and Interactions, Elsevier, Amsterdam, 2004, p. 215. [21] C. Stubenrauch, R. von Klitzing, J. Phys. Condens. Matter 15 (2003). [22] S.D. Taylor, J. Czarnecki, J. Masliyah, J. Colloid Interface Sci. 251 (2002) 149. [23] K.G. Marinova, T.D. Gurkov, T.D. Dimitrova, R.G. Alargova, D.K.W. Smith, Prog. Colloid Polym. Sci. 110 (1998). [24] O.D. Velev, T.D. Gurkov, R.P. Borwankar, J. Colloid Interface Sci. 159 (1993) 497. [25] A.V. Nguyen, H.J. Schulze, Colloidal Science of Flotation, Dekker, New York, 2004.