<p>Cationic starches are used on a large scale in paper industry as wet-end additives. They improve dry strength. retention of fines and fillers, and drainage. Closure of the white water systems in the paper mills hase increased the concentration of detrimental substances. This might be the reason for the poor retention of cationic starches observed in the last few years.<p>The purpose of the research described in this thesis was to obtain a better understanding of the adsorption of cationic starch on cellulose and how this interaction can be disturbed. In contrast to most research in papermaking we have used a colloid-chemical approach. This means that we kept our experimental system as simple as possible and therefore far from the reality of papermaking.<p>In chapter 2 we tried to generalize the specific problem of cationic starch adsorption on cellulose to polyelectrolyte adsorption on an oppositely charged surface. We used a recent extension of the polymer adsorption theory of Scheutjens and Fleer for polyelectrolyte adsorption to perform model calculations. It emerged that, for the adsorption of a strong polyelectrolyte on an oppositely charged surface, two regimes can be distinguished based on the effect of salt concentration on the adsorption. We call these the <em>screening-enhanced adsorption</em> regime and the <em>screening-reduced adsorption</em> regime. In the former regime the adsorption increases with increasing salt concentration because the repulsion between the segments is screened and a nonelectrostatic interaction between polyelectrolyte and surface is present. The adsorption decreases with increasing salt concentration in the latter regime, because the mainly electrostatic attraction between polyelectrolyte and surface is screened. A transition between these two regimes can take place if the balance between the electrostatic and nonelectrostatic interactions is changed. The electrostatic interactions are determined by the segment charge and the surface charge density. The strength of the nonelectrostatic interaction is described with a χs parameter, the net adsorption energy in units <em>kT.</em> From the model calculations it appears that the screening-reduced adsorption regime always occurs if the interaction between polyelectrolyte and surface is only electrostatic (χs=0). The polyelectrolyte can then be completely displaced from the surface with salt ions. If there is a nonelectrostatic attraction between polyelectrolyte and surface (χs>0) the <em>screening-enhanced</em> adsorption regime shows up in most cases. Only for very low segment charges and not too low surface charge densities, which is often the case for polyelectrolytes used in papermaking, we are dealing with the <em>screening-reduced adsorption</em> regime. The theory also predicts that the adsorbed amount shows a maximum as a function of the segment charge, irrespective of the value of χs. For a very low salt concentration this occurs at a segment charge of about 0.01 unit charges, or even lower.<br/>If the counter ions have a specific interaction with the surface, the adsorption of a polyelectrolyte can pass through a maximum as a function of the salt concentration provided χs is not too small.<br/>The predictions of the model calculations agree very well with experimental results reported in literature.<p>The careful characterization of the materials and the experimental methods we used are described in chapter 3. We showed that the microcrystalline cellulose, which we use as a model for cellulose fibers, is level-off DP cellulose. This means that we are dealing with fibers chemically cut into pieces. The microcrystalline cellulose had to be cleaned before use. because hemicellulose came off in aqueous solutions disturbing the determination of the equilibrium concentration of starch with a carbohydrate determination. This was accomplished by washing the microcrystalline cellulose with concentrated NaOH solutions. The surface charge of the microcrystalline cellulose, originating from carboxylate groups, was determined by potentiometric titrations to be about -1 C/g at pH=7, which is a little lower than reported for cellulose fibers. The specific surface area is a somewhat problematic quantity for a porous substrate as microcrystalline cellulose. Based on the adsorption of cationic polyelectrolytes with different molecular weights on microcrystalline cellulose, we estimated the accessible surface area for cationic starch to be about 6 m <sup>2</SUP>/g, which is only 10% of the surface area accessible to small ions.<p>We used two different types of cationic starch, namely cationic potato starch and cationic waxy maize starch. Potato starch consists of two components, an essentially linear polymer of α-1,4 glucose, called amylose, and a much larger branched polymer, called amylopectin. The fraction of amylose is about 21%. Waxy maize starch consists of amylopectin. only. From their sedimentation coefficients we estimated the molecular weight of cationic amylose to be about 3.5.10 <sup>5</SUP>and of cationic amylopectin from potato starch between 5.10 <sup>7</SUP>and 5.10 <sup>8</SUP>. The molecular weight of cationic amylopectin from waxy maize was estimated to be between 1.10 <sup>7</SUP>and 6.5.10 <sup>7</SUP>. Both cationic starches showed a marked decrease in viscosity and hydrodynamic radius, as measured by dynamic light scattering, with increasing electrolyte concentration. This indicates that cationic amylopectin has enough flexibility to shrink, even though it has a branched structure.<br/>Special attention is paid to the methods with which the starch concentration can be determined, especially to the well-known iodine determination. It is shown that it is very important to specify the iodine and iodide concentrations in the final solution with the blue starch-iodine complex and the wavelength at which the absorbance is measured.<br/>Finally we describe how the adsorbed amounts are measured by depletion.<p>In chapter 4 we investigated the adsorption of cationic amylopectin (from waxy maize, DS(Degree of Substitution) =0.035) on microcrystalline cellulose in the presence of simple electrolytes and at different pH values. The adsorption isotherms of cationic amylopectin were all of the high affinity type, as is expected for polyelectrolyte adsorption. The plateau value of the adsorbed amount showed a maximum as a function of the salt concentration. It was also found that the adsorbed amount, in the region<br/>where it decreases with increasing salt concentration, was very sensitive to the type of cation used. We obtained a Iyotropic series for the alkali cations, where the adsorbed amount in the presence of the cations decreased as Li <sup>+</SUP>>Na <sup>+</SUP>=K <sup>+</SUP>>Cs <sup>+</SUP>. The trend of these experimental results could be explained very well with the theory on polyelectrolyte adsorption described in chapter 2.<p>The plateau value of the adsorbed amount increased with increasing pH in the same way as the surface charge. The adsorbed amount of charge was estimated to be 10% of the titratable surface charge.<p>Based on the dependence of the adsorption on the pH and the salt concentration, we concluded that for cationic amylopectin charge interactions are the main driving forces for adsorption on cellulose.<p>The adsorption of cationic potato starch (usually DS=0.035) on microcrystalline cellulose is investigated in chapter 5. Special attention is paid to the fact that cationic potato starch is a mixture of 21% amylose and 79% amylopectin. It was found that amylose adsorbs preferentially. This was attributed to a larger accessible surface area for amylose due to its ability to enter the pores of microcrystalline cellulose during the equilibration time (15 hours).<p>The adsorption isotherms of cationic potato starch are also of the highaffinity type. There is a strong dependence on the cellulose concentration, caused by heterodispersity of the amylose and the amylopectin fractions.<p>The adsorption of cationic potato starch strongly and monotonously decreased with increasing salt concentration. It was completely displaced by salt ions at concentrations larger than 0.05 M. The divalent cations Ca <sup>2+</SUP>and Mg <sup>2+</SUP>appeared to be ten times as effective as Na+ in suppressing the adsorption of cationic potato starch, which is due both to their higher charge and a specific interaction with the cellulose surface. From the small difference in effect of Ca <sup>2+</SUP>and Mg <sup>2+</SUP>we concluded that the phosphate groups in cationic Potato starch play no relevant role in the adsorption.<p>Increasing the pH led to increasing adsorption. The adsorbed amount of charge was estimated to be 10% of the titratable charge.<br/>Finally, we investigated the effect of DS on the adsorption of cationic potato starch. At 2 mM NaCl the adsorbed amount of the starch with the lowest DS (0.017) was largest, but at 10 mM NaCl the difference between starches with DS=0.017, 0.035 and 0.047 was very small. The adsorbed amounts decreased slightly with decreasing DS. Theory on polyelectrolyte adsorption predicts that at a salt concentration of about 0.01 M the adsorbed amounts of polyelectrolytes with various segment charges can be the same indeed. The effect of segment charge is larger at lower and higher salt concentration. At a low salt concentration the starch with the lowest DS is expected to adsorb most, whereas at high salt concentration the starch with the highest DS will adsorb best.<br/>We concluded from the strong dependence on salt concentration and pH that the adsorption of cationic potato starch on cellulose is mainly driven by electrostatic attraction.<p>In chapter 6 we conclude that the adsorption of cationic starch on cellulose is mainly determined by the presence of charges and not by certain special properties of starch and cellulose. We point out that this thesis has general relevance because of the new light it sheds on polyelectrolyte adsorption. For papermaking this thesis is particularly relevant, because it explains the adsorption behaviour of cationic starch as that of polyelectrolytes and it therefore also improves the understanding of the adsorption of other polyelectrolytes used in papermaking. The adsorption behaviour of cationic potato starch is compared with that of cationic amylopectin from waxy maize. We suggest that this may be caused by differences in size and shape. Finally we indicate which results can be of direct relevance for papermaking in practice.