The composition and stability of polymer-magnetite complexes is essential for their use as a treatment for retinal detachment, for drug targeting and delivery, and for use as a MRI contrast agent. This work outlines a general methodology to design well-defined, stable polymer-magnetite complexes. Colloidal modeling was developed and validated to describe polymer brush extension from the magnetite core. This allowed for the observation of deviations from expected behavior as well as the precise control of polymer-particle complex size. Application of the modified Derjaguin-Verwey-Landau-Overbeek (DLVO) theory allowed the determination of the polymer loading and molecular weight necessary to sterically stabilize primary magnetite particles. Anchoring of polyethers to the magnetite nanoparticle surface was examined using three different types of anchor groups: carboxylic acid, ammonium, and zwitterionic phosphonate. As assessed by dynamic light scattering (DLS), the zwitterionic phosphonate group provided far more robust anchoring than either the carboxylic acid or ammonium anchor groups, which was attributed to an extremely strong interaction between the phosphonate anchor and the magnetite surface. Coverage of the magnetite surface by the anchor group was found to be a critical design variable for the stability of the zwitterionic phosphonate groups, and the use of a tri-zwitterionic phosphonate anchor provided stability in phosphate buffered saline (PBS) for a large range of polymer loadings. Incorporation of an amphiphlic poly(propylene oxide)-b-poly(ethyelene oxide) (PPO-b-PEO) diblock copolymer attached to the magnetite surface was examined through colloidal modeling and DLS. The relaxivity of the complexes was related to aggregation behavior observed through DLS. This indicated the presence of a hydrophobic interaction between the PPO layers of neighboring complexes. When this interaction was large enough, the complexes exhibited an increased relaxivity and cellular uptake. Thus, we have developed a methodology that allows for design of polymer-magnetite complexes with controlled sizes (within 8% of predicted values). Application of this methodology incorporated with modified DLVO theory aids in the design of colloidally stable complexes with minimum polymer loading. Finally, determination of an anchor group stable in the presence of phosphate salts at all magnetite loadings allows for the design of materials with minimum polymer loadings in biological systems. / Ph. D.

There is a need for multifunctional polymer-particle complexes for use in biomedical applications such as for drug delivery or as MRI contrast agents where composition and stability are essential for the complexes to function. This work outlines a general methodology for rationally designing complexes stabilized with polymer brush layers using adapted star polymer models for brush extension and pair potential. Block copolymer micelles were first utilized for experimental validation by using the brush extension model to predict the size and the interaction model to predict the second virial coefficient, A2. Subsequently, the models were used to predict the size and colloidal stability of magnetite-polymer complexes using the modified Deryaguin-Verwey-Landau-Overbeek theory. Novel hydrophilic triblock copolymers comprised of poly(ethylene oxide) tailblocks and a carboxylic acid containing polyurethane center block were examined by static and dynamic light scattering (SLS and DLS), small angle neutron scattering (SANS), and densiometry. Under conditions when the charge is suppressed such as at low pH and/or high ionic strength, the polymer chains self-assemble into micelles, whereas unimers alone are present under conditions where charge effects are important, such as high pH and low ionic strength. A model for effective interaction between star polymers was used to obtain an expression for the second virial coefficient (A2) for micelles in solution. The values of A2 obtained using this method were compared with experimentally determined values for star polymers and micelles. In doing so, not only was a new means of calculating A2 a priori introduced, but the applicability of star polymer expressions to micellar systems was established. Through the analogy of micelles to sterically stabilized nanoparticles, this model was applied to water-soluble block copolymers adsorbed on magnetite nanoparticles for the purpose of tailoring a steric stabilizing brush layer. The sizes of the magnetite-polymer complexes were predicted using the star polymer model employed for the micelle study with an added layer to account for the anchor block. Colloidal stability was predicted from extended DLVO theory using the pair interaction. This work will lead to a better understanding of how to design ion-containing block copolymers for steric stabilization of metal oxide nanoparticles. / Ph. D.