Inference of polymer adsorption from electrical double layer measurements : the silver iodide-polyvinyl alcohol system

The purpose of this study was to investigate how the double layer properties of charged particles are modified by the presence of adsorbed polymer molecules and to obtain information on the conformation of the polymer layer from the observed alterations in the double layer properties.In chapter 1. the use of double layer investigations to obtain insight in the adsorbed layer conformation is briefly outlined. Some theoretical and experimental aspects of the studies of polymer adsorption are shortly reviewed.For the experiments the AgI-PVA system is chosen. Double layer charge and potential of AgI dispersed in aqueous electrolytes can be determined and controlled. Much is known about the surface area determination and the stability of Agl. PVA is a water soluble, flexible and uncharged polymer of which the concentration in solution can be determined readily. The combined AgI-PVA system is well suited for the purpose of this study. For the interpretation of the results recourse can be made to similar information previously obtained with low molecular weight alcohols.The general procedures and the preparation of the Agl precipitates and sols are given in chapter 2.Chapter 3. deals with the characterization of the specific surface area of the AgI samples. Several independent methods are used: capacitance measurements, N2- adsorption, adsorption from solution and electron microscopy. Accepting an uncertainty margin of 10 to 20% the areas obtained by the last three methods compared well mutually. However, the capacitance areas were always 3 to 4 times greater. This disparity was observed before by VAN DEN HUL and LYKLEMA. Surface areas of AgI sols strongly reduce upon coagulation or precipitation. A subsequent heat treatment enhances this effect.We used the capacitance area in electrochemical studies, whereas for the adsorption of organic molecules the methylene blue adsorption area is chosen, otherwise unrealistically low adsorption values were found.In chapter 4. the properties and solution characteristics of PVA are described. The samples used, differing in molecular weight and acetate content, are characterized by IR and UV spectroscopy. It could be concluded that our samples are atactic, contain no or very little 1,2-glycol units and one or two conjugated carbonyl groups per molecule. The acetate groups in PVA 88 (12 mole % acetate groups) are predominantly distributed in blocks along the polymer chain, whereas in PVA 98 the distribution of these groups is probably random. Molecular weights and the molecular weight distributions were determined by viscometry and gel permeation chromatography.The solution properties of the polymers in water have also been studied by viscometry. Unperturbed dimensions, linear expansion factors and polymersolvent interaction parameters are calculated, taking heterodispersity into account. The relative magnitude of the steric hindrance or the characteristic ratio, combined with the fact that water is a poor solvent for PVA suggest that intramolecular interactions occur in the polymer chain.Chapter 5. covers the measurement of the mass of PVA adsorbed per m ² AgI, Γ <em>_p</em> , as a function of the PVA concentration. Special emphasis is. given to the influence of molecular weight, acetate content and of the surface charge and state of dispersion of the Agl. The adsorption isotherms show a high-affinity character, leading to a maximum amount adsorbed of 1.5 to 2.6 mg m ^-2 . The saturation adsorption increases with increasing molecular weight and acetate content. The surface charge of the Agl and its state of dispersion have no measurable influence on the adsorption. The reduction in adsorption by the addition of KNO ₃ UP to 10 ^-1 M is due to a decrease in available surface area.No desorption of the polymer could be detected upon dilution with solvent, but the increase in adsorption with time shows that the adsorbed segments are reversibly bound.In chapter 6. the principles of the double layer investigations are explained, whereafter a description is given of the potentiometric titrations and electrophoresis studies. The titrations reveal three important features upon adsorption of PVA:- the double layer capacitance decreases,- the p.z.c. moves to more positive values,- the curves pass through a common intersection point, characteristic for the type of PVA.These features reflect changes in the Stern-layer, caused by adsorbing polymer trains. Adsorbed polymer trains and low molecular weight adsorbates having a composition comparable with that of the polymer segments behave very similarly in the Stern-layer. From this resemblance it could be concluded that in PVA 88 segments with an acetate group adsorbed preferentially in the first layer. This is promoted by the blocky distribution of these groups in PVA 88. The similarity has further been used to develop a theorem to obtain the degree of occupancy of train segments in the Stern-layer, θ. The obtained result is confirmed by electrophoresis studies. A measure of the effective layer thickness can be found from the slope of the electrophoretic mobility against <em>pAg</em> curve around the isoelectric point. It was shown that this procedure is superiour to the classical one, in which the effective layer thickness is deduced from the reduction of the mobility in the plateau region.The main conclusion is that the fraction of polymer adsorbed in trains and the effective layer thickness are a function of the adsorbed amount only. Molecular weight and acetate content affect the adsorbed amount and thus indirectly influence the occupancy in the first layer and the effective thickness. Except for PVA 3-98, where probably preferential adsorption of higher molecular weight species occurs, the layer thickness is proportional to the square root of the degree of polymerization, as expected theoretically. The dependence of the train segment adsorption on the surface charge, though in principle present, istoo small to be practically important. The layer thickness is also independent of the surface charge.In chapter 7. some further parameters of the adsorbed layer are calculated, such as the fraction of segments adsorbed in trains, <em>p</em> , and the amount adsorbed in loops and tails. It can be concluded that molecules adsorbed in the isolated state (Γ <em>_p</em> ->0) <em></em> do not completely unfold, probably due to the presence of intra molecular interactions. The average segment density in the loop region is relatively high, indicating that the average loop length is much shorter than the total chain length.The obtained adsorption parameters are used for a quantitative check of the Hoeve model for polymer adsorption. The adsorbed amount at large polymer concentration is reasonably well predicted, including the molecular weight dependence. Also the trends in <em>p</em> (Γ <em>_p</em> ) <em></em> and θ(Γ <em>_p</em> ) were confirmed. However, the adsorbed layer thickness is very poorly predicted and the physical significance of the model parpameters is obscure. Our general conclusion therefore is that the Hoeve model cannot fully describe the adsorption of AgI on PVA.In conclusion, this study shows that double layer investigations combined with polymer adsorption measurements provide a valuable tool to investigate the conformation of adsorbed polymers. The degree of occupancy of the first layer, the effective thickness of the adsorbed layer and the kind of segments directly adsorbed onto the surface could be determined over a wide adsorption range.