7964
Langmuir 2006, 22, 7964-7968
Articles Effect of Binding of an Oligomeric Cationic Fluorosurfactant on the Dilational Rheological Properties of Gelatin Adsorbed at the Air-Water Interface Ashwin Rao,*,†,| Yongsin Kim,‡ Charles M. Kausch,‡ and Richard R. Thomas*,‡ Department of Polymer Science, The UniVersity of Akron, Akron, Ohio 44325-3909, and OMNOVA Solutions Inc., 2990 Gilchrist Road, Akron, Ohio 44305-4418 ReceiVed April 10, 2006. In Final Form: June 1, 2006 The effect of binding of an oligomeric cationic fluorooxetane surfactant on the interfacial properties of adsorbed gelatin-fluorooxetane complexes has been studied using dynamic surface tension and dilational rheological measurements. Adsorption kinetics of gelatin-fluorooxetane complexes are reminiscent of a mixed (barrier/diffusion limited) process, while the dilational rheological properties of the interface exhibit a strong dependence on surfactant concentration. At low surfactant concentrations, dilational surface moduli as well as phase angles are relatively insensitive to the presence of the fluorooxetane. However, at the critical aggregation concentration of the polymersurfactant system, there is a sharp increase in the complex modulus. Further increase in the fluorooxetane concentration does not significantly affect the complex modulus. The phase angle, however, does increase with increasing fluorooxetane concentration due to the transport of bound fluorooxetane from the subsurface to the solution-air interface. These results indicate that, at fluorooxetane concentrations exceeding the critical aggregation concentration, the polymersurfactant complexes adsorb to form cross-linked multilayers at the solution-air interface.
Introduction Viscoelastic properties of adsorbed polymer-surfactant complexes play a crucial role in determining the stability of foams and emulsions in the food industry and in coating processes used to manufacture multilayer films.1-3 These properties are a function of the chemistry and structure of adsorbed polymer-surfactant complexes. For example, elasticity of an interface containing adsorbed polymer chains has been attributed to the existence of physical entanglements as well as to the presence of hydrophobic interactions between adsorbed chains.4 Electrostatic interactions between adsorbed chains can enhance interfacial elasticity by promoting network formation between oppositely charged groups.5 Since surfactant binding to polymer chains results in changes to structure and conformation of molecules within the adsorbed layer, viscoelastic properties of the adsorbed layer are strongly dependent on the concentration of bound surfactants. Additionally, the presence of the surfactant molecules can also influence the nature of inter- and/or intramolecular interactions within the adsorbed layers. For example, when bound surfactant leads to an enhancement in the hydrophobic character of the * To whom correspondence should be addressed. E-mail:
[email protected] (A.R.);
[email protected] (R.R.T.). † The University of Akron. ‡ OMNOVA Solutions Inc. | Current address: Polymers Division, National Institute of Standards and Technology, 100 Bureau Drive, Stop 8500, Gaithersburg, MD 208998500. (1) Miller, R.; Wu¨stneck, R.; Kra¨gel, J.; Kretzschmar, G. Colloids Surf., A 1996, 111, 75. (2) Bos, M. A.; van Vliet, T. AdV. Colloid Interface Sci. 2001, 91, 437. (3) Miller, R.; Fainerman, V. B.; Makievski, A. V.; Kra¨gel, J.; Grigoriev, D. O.; Kazakov, V. N.; Sinyachenko, O. V. AdV. Colloid Interface Sci. 2000, 86, 39. (4) Noskov, B. A.; Loglio, G.; Miller, R. J. Phys. Chem. B 2004, 108, 18615. (5) Rao, A.; Kim, J.; Thomas, R. R. Langmuir 2005, 21, 617.
adsorbed layer, the adsorbed polymer-surfactant complexes can form interfacial microgels.6-9 However, if the effect of surfactant binding is a disruption of network structure,10 then a drop in elastic properties of the interface is observed.5 In this paper, interfacial dilational rheological measurements of gelatin solutions containing an oligomeric cationic fluorooxetane are reported. Gelatin is an ampholytic polymer with an isoelectric point (IEP) that varies between 7.5 and 9.3 for Type A and 4.7-5.2 for Type B gelatin.11,12 The composition of gelatin can be represented as -(Gly ∼ 33 - X ∼ 54 - (H)Pro ∼ 13)-, where Gly is the amino acid residue glycine, X includes small fractions of many amino acids, and Pro and HPro are proline and hydroxyproline residues, respectively, in a molar ratio ≈1.3:1, respectively.11,12 When the pH exceeds the IEP, amino acid residues such as aspartic acid (pKa ) 4.5), glutamic acid (pKa ) 4.5), and histidine groups (pKa ) 6-7.5) of gelatin are charged negatively while groups such as lysine (pKa ) 10), arginine (pKa ) 12), and tyrosine (pKa ) 12) are charged positively. Consequently, by varying the pH of gelatin solutions, it is possible to study behavior of charged polymer-surfactant systems ranging from anionic surfactant/cationic-polymer to cationic surfactant/ anionic-polymer mixtures. Additionally, the presence of nonionic hydrophobic residues on the gelatin chain, such as leucine, (6) Monteux, C.; Williams, C. E.; Meunier, J.; Anthony, O.; Bergeron, V. Langmuir 2004, 20, 57. (7) Monteux, C.; Fuller, G. G.; Bergeron, V. J. Phys. Chem. B 2004, 108, 16473. (8) Monteux, C.; Llauro, M. F.; Baigl, D.; Williams, C. E.; Anthony, O.; Bergeron, V. Langmuir 2004, 20, 5358. (9) Monteux, C.; Williams, C. E.; Bergeron, V. Langmuir 2004, 20, 5367. (10) Cooke, D. J.; Dong, C. C.; Thomas, R. K.; Howe, A. M.; Simister, E. A.; Penfold, J. Langmuir 2000, 16, 6546. (11) Djabourov, M. Contemp. Phys. 1988, 29, 273. (12) Wu¨stneck, R.; Kra¨gel, J. In Characterization of Gelatin/Surfactant Interaction and its ReleVance to Liquid Film Coating; Mo¨bius, D., Miller, R., Eds.; Elsevier Science: New York, 1998; Vol. 7, p 433.
10.1021/la060971d CCC: $33.50 © 2006 American Chemical Society Published on Web 08/17/2006
Oligomeric Cationic Fluorosurfactant Binding Effect
Langmuir, Vol. 22, No. 19, 2006 7965
Figure 1. Structure of cationic oligo(fluorooxetane) surfactant.
isoleucine, valine, and methionine, promotes binding of hydrophobic surfactant tails to gelatin molecules. In the case of the gelatin-fluorooxetane system, the binding of surfactant molecules should be determined by a combination of factors including electrostatic (attractive as well as repulsive) and hydrophobic interactions between the two species including conformational constraints arising due to size of the two components. Experimental Section Materials. Gelatin was Type B photographic grade (CL40, alkaliprocessed, ossein, isoelectric point ∼ 5) and provided by Eastman Kodak Company. Gelatin molecular weight was determined by gel permeation chromatography (GPC; phosphate-buffered pH 7 water at 60 °C by American Polymer Standards Corporation, Mentor, OH 44061-0901) using poly(ethylene glycol) standards. The following values were found: Mn ) 57000, Mw ) 139000, Mz ) 246000, and Mw/Mn ) 2.44. Water was distilled before use. The fluorosurfactant used in this study is an oligomeric cationic fluorooxetane whose structure is shown in Figure 1. Details of the synthesis, characterization, and interfacial properties of this material have been published elsewhere.13 Sample Preparation. Surfactant solutions were prepared by dissolving the fluorooxetane in distilled water at room temperature. Gelatin solutions (7 wt %) were prepared, at their natural pH (∼6.1), by adding the desired amount of dry gelatin to distilled water at room temperature and allowing the gelatin to swell for 10 min. The solution was heated to 60 °C with stirring until the gelatin was dissolved. The temperature was then reduced to 40 ( 1 °C where all measurements were performed. At the above temperature and pH, gelatin molecules adopt a random coil conformation and have a net negative charge. For surfactant-gelatin mixtures, the aforementioned procedure to prepare the gelatin solution was performed with the addition of the appropriate amount of a fluorosurfactant stock solution. No attempts were made to stabilize the gelatin solutions against microbiotic decomposition that occurs within 24 h at room temperature. Fresh solutions were prepared for each experiment. Instrumentation. Equilibrium and dynamic surface tensions and interfacial rheological measurements were performed using an oscillating bubble rheometer (The Tracker, ThetaDyne Corporation). Details of the instrumentation have been published previously.5 Solutions were maintained at 40 ( 1 °C using a Neslab RTE-140 circulating bath. A bubble was formed inside the solution to be examined using an inverted needle. Images of the bubble profile were captured, digitized, and analyzed by the Laplace equation to determine surface tensions. Interfacial rheological measurements were performed by oscillating the volume of the bubble ((20% of the initial volume) and determining the viscoelastic moduli from the change in the surface tension. The frequency of the oscillations was varied from 0.03 to 1 Hz. Interfacial Dilational Rheology. In a two-dimensional dilational rheological measurement, the area of an interface is changed while ensuring that the shape of the interface remains constant.2 The viscoelastic properties of the interface are determined from the corresponding change in interfacial tension during the deformation process.1,14,15 For a sinusoidal variation in area of the interface (having (13) Rao, A.; Kim, Y.; Kausch, C. M.; Russell, V. M.; Thomas, R. R. Langmuir 2006, 22, 4811. (14) Cao, X.; Li, Y.; Jiang, S.; Sun, H.; Cagna, A.; Dou, L. J. Colloid Interface Sci. 2004, 270, 295. (15) Casca˜o Pereira, L. G.; The´odoly, O.; Blanch, H. W.; Radke, C. J. Langmuir 2003, 19, 2349.
Figure 2. Dynamic surface tension of pure gelatin (A), pure fluorooxetane (B), and gelatin-fluorooxetane mixtures (C) at the air-solution interface. The surfactant concentrations used to obtain the dynamic surface tension plots are 4.9 × 10-6 M (0), 6.27 × 10-6 M (O), 2.45 × 10-5 M (3), 4.08 × 10-5 M (4), 8.7 × 10-5 M (]), 1.63 × 10-4 M (open left-facing triangle), and 1.87 × 10-4 M (open right-facing triangle). The solid lines are a guide to the eye. an amplitude Aa, an equilibrium value of Ao, and a frequency ω) represented by the equation, ∆A ) A - A0 ) Aa sin(ωt)
(1)
the change in surface tension can be described as ∆γ ) γ - γ0 ) γa sin(ωt + φ)
(2)
where γa is the measured amplitude of the surface tension, γ0 is the equilibrium surface tension, and φ is the phase angle. The absolute value of the complex surface dilational modulus is then defined as |E| )
γa Aa/A0
(3)
that can be decoupled into real and imaginary components E′ and E′′ by the following equations E′ ) |E| cos φ E′′ ) |E| sin φ
(4)
where E′ is the storage modulus representing the elastic properties of the adsorbed layer and E′′ is the loss modulus arising due to transport-induced disruption of surface tension gradients during the dilational process.
Results and Discussion Dynamic surface tension measurements were used to characterize the adsorption kinetics at the air-water interface. The adsorption kinetics of pure gelatin indicate a slow equilibration of the surface tension (Figure 2A). Such a response, typical of protein adsorption at the air-water interface, occurs due to slow rearrangement of gelatin molecules upon adsorption at the interface. The adsorption mechanism of fluorooxetane molecules at the air-water interface is a diffusion-limited process and,
7966 Langmuir, Vol. 22, No. 19, 2006
Figure 3. Adsorption isotherm for pure fluorooxetane (O) (A) and the 7 wt % gelatin-fluorooxetane mixtures (0) (B) at 40 °C. The dashed line is a guide to the eye while the solid line is a fit to the Davies adsorption isotherm equation. The critical micelle concentration of the fluorooxetane solutions (cmc) and the critical aggregation concentration (cac) of the gelatin-fluorooxetane mixtures are also identified.
except for the lowest surfactant concentration16 used in this study, involves a rapid equilibration of surface tension (Figure 2B). Finally, in Figure 2C, adsorption kinetics of gelatin-fluorooxetane mixtures are illustrated. The lower values of surface tension of gelatin-fluorooxetane mixtures, as compared to those of pure gelatin, indicate the presence of fluorooxetane at the interface. At the lowest surfactant concentration, adsorption kinetics of the gelatin-fluorooxetane mixture are similar to those observed for pure gelatin molecules. With increasing fluorooxetane concentration, the change in dynamic surface tension occurs at a faster rate. However, the adsorption kinetics are still representative of a mixed (barrier/diffusion) process, indicating that the species adsorbing at the interface are a complex comprised of gelatin and fluorooxetane molecules. The increase in adsorption rate with increasing fluorooxetane concentration may be a result of hydrophobic interactions between gelatin and surfactant molecules that leads to the formation of a compact gelatin-fluorooxetane complex. The adsorption isotherms of the fluorooxetane as well as gelatin-fluorooxetane mixtures are shown in Figure 3. The critical micelle concentration (cmc) of fluorooxetane in water was estimated to be 1.33 × 10-4 mol/L. In the case of fluorooxetane, adsorption isotherm data were fitted to the Davies adsorption isotherm equation,17 assuming a pseudo-single surfactant approach,18 and at saturation yielded a surface excess ≈(3.39 ( 0.08) × 10-6 mol/m2 corresponding to a molecular area at saturation ≈49.0 ( 1 Å2/molecule. Pure gelatin solutions (7 wt %) have a surface tension ≈38 mN/m after a long equilibration time (∼6 h). With the addition of fluorooxetane, there is a decrease in surface tension due to adsorption of fluorosurfactant molecules at the interface. The adsorption isotherm of the gelatinfluorooxetane mixture shows the existence of a plateau region as the fluorooxetane concentration increases from 4 × 10-5 to 9 × 10-5 M. The onset of the plateau region corresponds to the critical aggregation concentration (cac) of the fluorooxetane while the end of the plateau region occurs when gelatin molecules are saturated with surfactant molecules. The relatively low concentration regime over which the plateau in the surface tension occurs suggests weak binding between gelatin chains and cationic (16) Surface tension measurements at very dilute surfactant concentrations can contain contributions from surface-active impurities present in the solution and should be interpreted with caution. (17) Davies, J. T. Proc. R. Soc. London, Ser. A 1958, 245, 417. (18) Daniel, R. C.; Berg, J. C. Langmuir 2002, 18, 5074.
Rao et al.
Figure 4. Complex surface dilational moduli for pure fluorooxetane (A) and 7 wt % gelatin-fluorooxetane mixtures (C) along with phase angles for pure fluorooxetane (B) and a 7 wt % gelatin/ fluorooxetane mixture (D) as a function of added surfactant concentration and frequency at 40 °C. Shown are the complex surface moduli E* (A and C) and phase angle (B and D) for frequencies 0.033 (0), 0.05 (O), 0.1 (4), 0.25 (3), and 0.5 (]) Hz. The dashed lines are the respective frequency-averaged moduli and phase angles for a pure 7 wt % gelatin solution at 40 °C. The solid lines are a guide to the eye.
fluorooxetane molecules. This observation is not surprising since cationic surfactants are known to have weaker interactions with gelatin compared to their anionic counterparts. However, in the present system, weak binding of fluorooxetane molecules to gelatin might also be aggravated due to a combination of factors related to the increased size of the fluorooxetane that may impose steric constraints on the binding of surfactant headgroups to anionic sites on the gelatin chain as well as conformational changes induced in the gelatin chain due to hydrophobic interactions between gelatin and fluorinated side groups of fluorooxetane chains. Interfacial dilational rheological properties of the fluorooxetane as well as gelatin-fluorooxetane mixtures are shown in Figure 4. The rheological behavior of fluorooxetane adsorbed at the air-water interface (Figures 4A and 4B) has been studied previously13 and can be described in the framework of the Lucassen-van den Tempel model.19,20 At low surface coverages, the response of the interface is predominantly elastic in nature and is characterized by low values of the phase angle. This elastic character arises due to the strong hydrophobic interactions between adsorbed fluorooxetane molecules. As the concentration of fluorooxetane in solution increases, the response of the adsorbed layer acquires a viscoelastic character due to transport of surfactant molecules from the bulk to the interface. This transport mechanism, indicating the presence of a soluble monolayer, reduces the modulus of the adsorbed layers, and eventually, the rheological response of the interface is characterized by low values of the storage modulus and a large phase angle. The rheological response of gelatin molecules adsorbed at the air-water interface is predominantly elastic in nature (dotted line in Figures 4C and 4D). This behavior, associated with an “insoluble” surfactant, is typical of the response exhibited by irreversibly adsorbed protein molecules at the air-water interface.15 The elastic character of the adsorbed gelatin layer arises due to the formation of a cross-linked network within adsorbed gelatin layers. The physical origin of cross-links can (19) Lucassen, J.; van den Tempel, M. Chem. Eng. Sci. 1972, 27, 1283. (20) Lucassen, J.; van den Tempel, M. J. Colloid Interface Sci. 1972, 41, 491.
Oligomeric Cationic Fluorosurfactant Binding Effect
Figure 5. Frequency dependence of the complex storage modulus (A, C, and E) and phase angle (B, D, and F) of pure gelatin (4), pure fluorooxetane (0), and gelatin-fluorooxetane (O) layers adsorbed at the solution-air interface at fluorooxetane concentrations of 2.45 × 10-5 M (E and F), 5.2 × 10-5 M (C and D), and 1.63 × 10-4 M (A and B).
be attributed to a combination of electrostatic, hydrogen bonding, and hydrophobic interactions between the different chemical groups present on the gelatin chain.21 The complex modulus and phase angle of adsorbed gelatinfluorooxetane complexes at fluorooxetane concentrations lower than the cac are shown in Figures 4C and 4D, respectively, while the frequency dependency of the complex storage modulus and phase angles at a fluorooxetane concentration of 2.45 × 10-5 M are shown in Figures 5E and 5F, respectively. This frequency dependence is representative of a general trend observed for all fluorooxetane concentrations below the cac. As seen in Figures 4C and 4D, the complex modulus and phase angle, respectively, of the adsorbed gelatin-fluorooxetane complexes have values that are slightly larger than those of a pure gelatin adsorbed layer but lower than pure fluorooxetane. Additionally, unlike surface tension, these values are not sensitive to a change in composition at the interface. This result is not unexpected since, at low fluorooxetane concentrations, gelatin molecules comprise the majority of adsorbed interfacial molecules. The results also suggest that the presence of fluorooxetane molecules at the interface does not have a significant affect on network formation between adsorbed gelatin molecules. The slight increase in storage modulus of a mixed gelatin-fluorooxetane layers may be attributed to the hydrophobic interactions betwen the gelatin and fluorooxetane molecules present in the adsorbed layer. A schematic of the adsorbed layer in this concentration regime is shown in Figure 6A. As the fluorooxetane concentration approaches the cac, there is a change in the viscoelastic nature of the adsorbed layer. The frequency dependence of complex modulus and phase angle of gelatin-fluorooxetane mixtures for one fluorooxetane concentration in the plateau region is plotted in Figures 5C and 5D, respectively. The absolute values of complex elastic modulus of the mixture are larger than those of pure gelatin or pure fluorooxetane layers while the phase angle of the mixture has intermediate values between the two pure systems. The effect of increasing fluorooxetane concentration beyond the cac on the phase angle and storage modulus is seen in Figures 4C and 4D, respectively. At a fluorooxetane concentration corresponding to the onset of the plateau in surface tension, there is a sudden increase in the complex modulus. Increasing the fluorooxetane (21) Wu¨stneck, R. Colloid. Polym. Sci. 1984, 262, 821.
Langmuir, Vol. 22, No. 19, 2006 7967
Figure 6. Figure 6A represents the structure of the gelatinfluorooxetane adsorbed layer at fluorooxetane concentrations that are lower than the cac. The green fluorooxetane molecules do not significantly affect the structure at the interface. Figure 6B represents the structure at the interface when the fluorooxetane concentration exceeds the cac. The gelatin chains at the surface are red in color while those in the underlying subsurface layer are colored blue. The fluorooxetane molecules are colored green. The interface is represented by the dotted line.
concentration beyond the cac does not result in a significant variation in the complex modulus. The second important result from Figure 4D is that the phase angle increases continuously with increasing fluorooxetane concentration, suggesting an increased influence of interfacial transport on the viscoelastic nature of the interface. The observed increase in values of the complex storage modulus at fluorooxetane concentrations exceeding the cac can occur due to two possible reasons: (i) an increase in the concentration of adsorbed molecules22-24 or (ii) an increase in network formation at the interface. Indeed, recent studies of the structure of anionic polyelectrolyte-cationic surfactant complexes adsorbed at the air-solution interface have found the existence of multilayers at surfactant concentrations exceeding the cac.25,26 The gelatin-fluorooxetane complexes may adopt a similar multilayered structure at the solution-air interface, a schematic of which is depicted in Figure 6B. Interfacial layering in polyelectrolyte-surfactant complexes has been associated with the formation of a gellike structure at the interface.8 However, dilational rheological measurements have been found to be insensitive to the existence of these structures.7 One explanation for this observation was the inhibition of surface tension gradients due to relatively rapid adsorption kinetics of the polymer-surfactant complex. As seen in Figure 2C, the adsorption of the gelatin-fluorooxetane complexes occurs on a time scale that is longer than the slowest deformation rate employed in this study. It is this feature of the adsorption process that leads to dilational measurements being sensitive to the multilayered structure of adsorbed molecules. The continuous increase in measured values of phase angle of the adsorbed layer indicates the existence of a transport (22) Hempt, C.; Lunkenheimer, K.; Miller, R. Z. Phys. Chem. 1985, 266, 713. (23) Wu¨stneck, R.; Kretzschmar, G.; Zastrow, L. Kolloid. Zh. 1987, 49, 207. (24) Wu¨stneck, R.; Krotov, V. V.; Ziller, M. Colloid. Polym. Sci. 1984, 262, 67. (25) Taylor, D. J. F.; Thomas, R. K.; Penfold, J. Langmuir 2002, 18, 4748. (26) Zhang, J.; Thomas, R. K.; Penfold, J. Soft Matter 2005, 1, 310.
7968 Langmuir, Vol. 22, No. 19, 2006
mechanism at the interface. However, given the slow adsorption kinetics of the polymer-surfactant complexes, the mechanism of transport of surface-active molecules from the bulk to the surface should not be a significant contributor to the viscoelasticity of the interface. Instead, the increase in phase angle probably arises due to the transport of fluorooxetane molecules from the subsurface layer to the interface.27 This mechanism exists due to weak binding of fluorooxetane molecules to gelatin that permits surfactant molecules to diffuse within the surface and subsurface layers during dilational measurements. The concentration dependence of the phase angle simply reflects an increase in concentration of fluorooxetane molecules proximal to the interface. Additionally, hydrophobic interactions between fluorooxetane molecules and gelatin chains may promote the formation of physical cross-link junctions between chains in adjacent layers, thus increasing surface elasticity. The plateau in the surface tension isotherm for the gelatinfluorooxetane system ceases at a fluorooxetane concentration of 9 × 10-5 M and the surface tension of the gelatin-fluorooxetane mixture quickly attains the values corresponding to that of a saturated fluorooxetane system. The rheological response of the adsorbed layers at a fluorooxetane concentration of 1.63 × 10-4 M is plotted in Figures 5A and 5B. Surprisingly, even at these concentrations;where the surface tension of the mixture is identical to that of the pure surfactant solution;the surface retains its elastic character, indicating that fluorooxetane molecules are unable to displace gelatin molecules from the solution-air (27) Wantke, K. D.; Ortegren, J.; Fruhner, H.; Andersen, A.; Motschmann, H. Colloids Surf., A 2005, 261, 75.
Rao et al.
interface. Due to the limited solubility of fluorooxetane in water, interfacial rheological measurements were not pursued beyond a maximum surfactant concentration of 1.8 × 10-4 M.
Conclusions Adsorption isotherms, dynamic surface tension, and interfacial rheological properties were measured for gelatin-fluorooxetane mixtures at 40 °C. The polymer-surfactant complexes exhibited adsorption kinetics representative of a mixed (diffusion/barrier) limited process. One consequence of the slow adsorption process is an absence of a transport-based dissipative mechanism at the surface. Instead, dilational measurements of viscoelastic properties of adsorbed layers suggest that fluorooxetane diffusion from the sublayer to the interface is the predominant mode of energy dissipation at the interface. Additionally, at fluorooxetane concentrations exceeding the cac, surfactant binding to gelatin leads to the formation of cross-linked multilayers that provide a strong elastic character to the interface. A comparison of results from the current study with earlier work regarding rheological properties of anionic polyelectrolyte-cationic surfactant systems reveals that subtle differences in the structure and adsorption kinetics of the polymer-surfactant complexes can result in significantly different dilational rheological properties of adsorbed layers. Acknowledgment. Gelatin was provided kindly by Dr. Michael Orem of Kodak. A.R. also gratefully acknowledges graduate fellowships from the University of Akron and OMNOVA Solutions Inc. LA060971D