Electroosmotic flow is widely used as a solution pumping method in numerous microfluidic applications. This type of flow has several advantages over other pumping techniques, such as the fast response time, the ease of control and integration in different microchannel designs. The flow utilizes the scaling of channel dimensions, which enhances the effects of the electrostatic forces to create flow in microchannels under an electrical body force. However, the electrostatic properties of the solution/wall material pairings are unique and must be experimentally measured. As a consequence, accurate knowledge about the electrostatic properties of the solution and wall material pairings is important for the optimal design of microfluidic devices using electroosmotic flow. Moreover, the introduction of new solutions and new channel materials for different applications is common in the microfluidics area. Therefore, any improvement on the experimental techniques used to examine the electrostatic properties of microchannels is beneficial to the research community. In this work, an improvement to the current-monitoring technique for studying the electrokinetic properties of microchannels is achieved by replacing the conventional straight channel design with a new Y-channel design. The errors from both the undesired pressure driven flow and solution electrolysis were addressed and significantly reduced. The new design offers high accuracy in finding the electrokinetic properties of microchannels. The experimental outcome from the new channel design is better compared to the outcomes of the straight channel, which helps in distinguishing the important electroosmotic pumping regions from the current-time plot. Moreover the time effectiveness in performing the experiments with the new channel design is better compared to that for the straight channel design. A modified analysis approach is also presented and validated for finding the electrokinetic properties from the outcomes of the current-monitoring technique, which is called the current-slope method. This approach is validated by comparing its findings with the results of the conventional length method. It was found for most situations that the discrepancy between the two methods, the current-slope and total length method, are within the uncertainty of the experimental measurements, thus validating the new analysis approach. In situations where it is hard to distinguish the start and end of solution replacement from the current-time plot of the current-monitoring technique, the current-slope method is advised. With the new design, different parametric studies of electroosmotic flow in PDMS based microchannels are estimated. At first the zeta potential of biological buffers are studied. Moreover the effect of continuous electroosmotic pumping, the chip substrate structure, and temperature on the average zeta potential of microchannels are examined. It was found that for air plasma treated PDMS microchannels the chip substrate material does not have an effect on the average zeta potential of the microchannels. The following chemical treatments are attempted with the aim of improving the surface and electrostatic properties of PDMS based microchannels: prepolymer additive with acrylic acid, extraction of PDMS, and both heat and plasma induced HEMA (Hydroxyethyl methacrylate) grafting on the surface of PDMS. Extensive characterization is performed with different experimental methods. The stability of the artificial hydrophilic properties of the PDMS microchannels with time was improved with both the extraction and HEMA grafting techniques. On the other hand, there was no evidence of any improvement in the zeta potential of microchannels with the surface treatments.

Electroosmotic flow is widely used as a solution pumping method in numerous microfluidic applications. This type of flow has several advantages over other pumping techniques, such as the fast response time, the ease of control and integration in different microchannel designs. The flow utilizes the scaling of channel dimensions, which enhances the effects of the electrostatic forces to create flow in microchannels under an electrical body force. However, the electrostatic properties of the solution/wall material pairings are unique and must be experimentally measured. As a consequence, accurate knowledge about the electrostatic properties of the solution and wall material pairings is important for the optimal design of microfluidic devices using electroosmotic flow. Moreover, the introduction of new solutions and new channel materials for different applications is common in the microfluidics area. Therefore, any improvement on the experimental techniques used to examine the electrostatic properties of microchannels is beneficial to the research community. In this work, an improvement to the current-monitoring technique for studying the electrokinetic properties of microchannels is achieved by replacing the conventional straight channel design with a new Y-channel design. The errors from both the undesired pressure driven flow and solution electrolysis were addressed and significantly reduced. The new design offers high accuracy in finding the electrokinetic properties of microchannels. The experimental outcome from the new channel design is better compared to the outcomes of the straight channel, which helps in distinguishing the important electroosmotic pumping regions from the current-time plot. Moreover the time effectiveness in performing the experiments with the new channel design is better compared to that for the straight channel design. A modified analysis approach is also presented and validated for finding the electrokinetic properties from the outcomes of the current-monitoring technique, which is called the current-slope method. This approach is validated by comparing its findings with the results of the conventional length method. It was found for most situations that the discrepancy between the two methods, the current-slope and total length method, are within the uncertainty of the experimental measurements, thus validating the new analysis approach. In situations where it is hard to distinguish the start and end of solution replacement from the current-time plot of the current-monitoring technique, the current-slope method is advised. With the new design, different parametric studies of electroosmotic flow in PDMS based microchannels are estimated. At first the zeta potential of biological buffers are studied. Moreover the effect of continuous electroosmotic pumping, the chip substrate structure, and temperature on the average zeta potential of microchannels are examined. It was found that for air plasma treated PDMS microchannels the chip substrate material does not have an effect on the average zeta potential of the microchannels. The following chemical treatments are attempted with the aim of improving the surface and electrostatic properties of PDMS based microchannels: prepolymer additive with acrylic acid, extraction of PDMS, and both heat and plasma induced HEMA (Hydroxyethyl methacrylate) grafting on the surface of PDMS. Extensive characterization is performed with different experimental methods. The stability of the artificial hydrophilic properties of the PDMS microchannels with time was improved with both the extraction and HEMA grafting techniques. On the other hand, there was no evidence of any improvement in the zeta potential of microchannels with the surface treatments.

A monitorização terapêutica tem sido descrita como valioso recurso clínico, na individualização do regime de dosagem, de acordo com a concentração do fármaco e/ou de seus produtos de biotransformação, em amostras de plasma ou soro, coletadas com base no contexto clínico e nos princípios da farmacocinética. Em razão da complexidade dos fluidos biológicos e da baixa concentração dos fármacos nestas matrizes, a etapa de preparo de amostra, extração, pré-concentração dos analitos e eliminação dos interferentes, têm sido requerida para o desenvolvimento de métodos cromatográficos com alta sensibilidade e seletividade analítica. A extração sortiva em barra de agitação (SBSE), recente técnica de preparo de amostra, baseia-se no equilíbrio de sorção do analito entre as fases: extratora (polidimetilsiloxano) e amostra aquosa. A fase extratora PDMS é a única disponível no mercado, o que tem limitado a sensibilidade e a seletividade analítica da técnica SBSE. O uso de polipirrol (PPY) como fase extratora está relacionado às diferentes interações de seus grupos funcionais (hidrofóbica, -, com o grupo funcional polar, troca iônica, ácido-básica, dipolo-dipolo e dipolo induzido-dipolo.) com os analitos. Neste trabalho, uma nova fase extratora SBSE, com o revestimento misto PDMS/PPY foi avaliada para análises de antidepressivos em amostras de plasma, por cromatografia líquida de alta eficiência. A otimização das variáveis SBSE: tempo e temperatura de extração, tempo de dessorção e pH da matriz biológica, baseada no equilíbrio de sorção dos analitos entre as fases: polimérica (PDMS/PPY) e fluido biológico, permitiu a determinação dos analitos em concentrações plasmáticas que contemplam o intervalo terapêutico. A presença de estrutura porosa (PPY) e não porosa (PDMS) na superfície polimérica da barra extratora SBSE-PDMS/PPY foi confirmada através de análises por Microscopia Eletrônica de Varredura (MEV). De acordo, com os resultados obtidos nas análises de MEV e nas análises (individuais e simultâneas) de amostras de plasma enriquecidas com os analitos, os mecanismos de retenção dos fármacos junto à superfície PDMS/PPY ocorreram através dos processos de adsorção (PPY) e absorção (PDMS). A validação analítica foi realizada segundo as normas da Agência Nacional de Vigilância Sanitária. O método padronizado apresentou linearidade no intervalo de concentração plasmática que variou dos limites de quantificação a 500 ng mL-1, coeficientes de determinação maiores que 0,994, precisão inter ensaios com coeficientes de variação menores que 15% e exatidão de 96% a 106%. Os valores de limites de quantificação obtidos são congruentes com a menor concentração plasmática do intervalo terapêutico preconizado. O método SBSE-PDMS/PPY padronizado e validado foi utilizado para determinações dos antidepressivos, sertralina, duloxetina e fluoxetina em amostras de plasma de pacientes, em terapia com estes fármacos. Desta forma, o método SBSE-PDMS/PPY poderá ser empregado para fins de monitorização terapêutica. / Therapeutic drug monitoring has been described a valuable clinical resource for the customization of the dosage regimen, in accordance with the drug concentration and/or its biotransformation products, in of plasma or serum samples, collected on the basis of clinical context and pharmacokinetics principles. Due the complexity of biological fluids and the low concentration of the drug in these matrices, an stage of sample preparation, extraction, pre-concentration of the analytes and elimination of the interferents has been required for the development of chromatographic methods with high sensitivity and analytical selectivity. Stir bar sorptive extraction (SBSE), a recent sample preparation technique, is based on the sorption equilibrium of the analytes between the polydimethylsiloxane (PDMS) and aqueous phases. PDMS is the only commercially available extraction phase (SBSE), which has limited the analytical sensitivity and the selectivity of the SBSE technique. The use of polypyrrole (PPY) as extraction phase is related to the different interactions of its functional groups (- interactions, polar groups interactions, acid-base, dipole-dipole, dipole-induced-dipole) and the analytes. In this work, a new SBSE extraction phase, with a PDMS/PPY coating was evaluated for the analysis of antidepressants in plasma samples by High Performance Liquid Chromatography. The optimization of the SBSE variables extraction time and temperature, dessorption time and pH of the biological matrix based on the sorption equilibrium of the analytes between the polymeric (PDMS/PPY) and biological fluid phases, allowed determination of the analytes in plasmatic concentrations that correspond to the therapeutic interval. The presence of a porous structure (PPY) as well as no none, porous (PDMS), on the polymeric surface of SBSE-PDMS/PPY was confirmed by Scanning Electron Microscopy (SEM) analyses. In agreement with the results obtained by SEM analyses and individual and simultaneous analyses of the plasma samples spiked with the analytes, the mechanisms of drugs retention on the surface of PDMS/PPY occur through adsorption (PPY) and absorption (PDMS). Analytical validation was carried through according to the norms of the National Agency of Sanitary Vigilance. The standardized method presented linearity in the plasmatic interval concentrations that varied from the limits of quantification to 500 ng mL-1, the determination coefficients were higher than 0.994, inter precision assays with coefficients of variation lower than 15%, and accuracy from 96% to 106%. The quantification value limits were in agreement with the lowest plasmatic concentration of the established therapeutical interval. The standardized and validated SBSE-PDMS/PPY method was used for determination of sertraline, duloxetine and fluoxetine in plasma patient samples under therapy with these drugs. Thus, the SBSE-PDMS/PPY method could be used for therapeutic drug monitoring.

Many of the materials used today for cardiovascular implants exhibit good bulk mechanical properties but fail to provide desirable surface properties for reducing thrombogenicity and promoting tissue integration. In fact, biological responses at the blood-material interface, including non-specific protein adsorption, coagulation, and platelet adhesion and activation significantly limit the use of currently available materials in many blood contacting applications. As our understanding of the biological responses to foreign materials has grown, so too has the potential for creating 'bioactive' materials capable of inducing and directing beneficial cellular processes. One promising technique for circumventing undesirable blood-biomaterial interactions involves seeding vascular endothelial cells (ECs) onto synthetic vascular grafts as a means of exploiting the physiological anticoagulant characteristics of the endothelium. Methods for improving cell retention on these constructs include immobilization of cell recognition motifs on the biomaterial surface in order to improve interactions between cells and the synthetic substrate. However, there remains the need to better understand the interactions between surface bound ligands and cells, and the role of linker molecule chemistry on ligand bioactivity and cellular response. In the current work, a novel method was optimized for modifying poly (dimethylsiloxane) (PDMS) with cell adhesion peptides tethered via a heterobifunctional allyl-, NSC-terminated polyethylene oxide (PEO) linker molecule. These novel surfaces combine the protein repellant property of PEO with the cell binding property of cell adhesion peptides. It was found that surfaces modified in this manner reduced protein adsorption to PDMS while increasing cell adhesion. Therefore the use of a generic PEO linker molecule was shown to be a very promising method of reducing non-specific protein interactions while maintaining ligand bioactivity. Silicone surfaces were also modified with diaminobutane (DAB) dendrimers in an attempt to increase the surface capacity for attachment of biomolecules and to compare the effect of surface peptide density with ligand mobility. Grafting cell adhesion peptides via surface bound dendrimers was found to increase the surface peptide density when compared to peptides grafted via the PEO spacer alone. However, cell adhesion was not significantly improved on the dendrimer-peptide modified surfaces compared to PDMS controls. This observation provides evidence that the properties of the linker molecule used for attachment of cell adhesion peptides to a biomaterial surface may be a critical factor in determining peptide bioactivity. In this case the peptides bound to the surface via the highly mobile linear PEO linker showed increased cell adhesion when compared to peptides linked via the rigid, highly branched dendrimer. It is therefore hypothesized that ligand mobility on a biomaterial surface may significantly influence ligand-cell receptor interactions to an even greater extent than surface peptide density. / Thesis / Master of Applied Science (MASc)

Doctor of Philosophy / Department of Chemistry / Christopher T. Culbertson / Microfluidic devices have a wide variety of biological applications. My Ph.D. dissertation focuses on three major projects. A) culturing a non-adherent immortal cell line within a microfluidic device under static and dynamic media flow conditions; B) designing and fabricating novel microfluidic devices for electrokinetic injecting analytes from a hydrodynamic fluid; and C) using this novel injection method to lyse single non-adherent cells by applying a high electric field across the cell at a microfluidic channel intersection. There are several potential advantages to the use of microfluidic devices for the analysis of single cells: First, cells can be handled with care and precision while being transported in the microfluidic channels. Second, cell culturing, handling, and analysis can be integrated together in a single, compact microfluidic device. Third, cell culturing and analysis in microfluidic devices uses only extremely small volumes of culturing media and analysis buffer. In this dissertation a non-adherent immortal cell line was studied under static media flow conditions inside a CO[subscript]2 incubator and under dynamic media flow conditions in a novel portable cell culture chamber. To culture cells they must first be trapped on a microfluidic device. To attempt to successfully trap cells, three different types of cellular traps were designed, fabricated and tested in polydimethylsiloxane (PDMS)-based microfluidic devices. In the first generation device, cubic-shaped traps were used. After 48 h of culturing in these devices the cell viability of 79 [plus or minus] 6 % (n = 3). In the second generation device, circular wells with narrow connecting channels were employed. However, after 12 h of culturing, no viable cells were found. While the second generation device was not capable of successfully culturing cells, it did demonstrate the importance of culturing under dynamic conditions which lead to next design. The third generation microfluidic device consisted of hydrodynamic shaped traps that were used to culture the cells in a less confined environment. The cell viability after 12 h in this design was 29 [plus or minus] 41% (n = 3). In addition to cell trapping, a novel electrokinetic injection method was developed for injecting analytes from a hydrodynamic flow into a separation channel that was followed by an electrokinetic separation. As the hydrodynamic flow could introduce some excess band broadening in the separation, the actual band broadening of an analyte was measured for different channel depths and hydrodynamic fluid flow rates. The results consistently showed that the separations performed on these devices were diffusion limited. Finally, using this novel injection method, single cell lysis was performed by applying a high voltage at the microfluidic channel intersection. The results of these studies may eventually be applied to help answer some fundamental questions in the areas of biochemistry and pharmaceutical science.

Fumed silica filler has long been used to structurally reinforce silicone elastomers. Unfortunately, the combination of as little as a few weight percent of untreated fumed silica nanoparticles [uFSN] with a siloxane polymer, such as PDMS, forms a difficult to process waxy solid admixture that even long periods of high shear mixing will not thin. In the course of the current work it was noted that after a period of storage certain solid admixtures would become viscous liquids when subjected to additional high shear mixing. It was further found that the required aging period could be decreased if the admixture storage temperature were increased. The only known interaction of PDMS and uFSN at moderate conditions is the adsorption of polymer on filler, and this interaction is also known to occur more quickly at higher temperature. This study examines the relationship between polymer adsorption and admixture liquefaction. Further, the mechanical properties of cured elastomers containing liquefied admixtures are examined to assess the degree of reinforcement that these materials afford.

Human surrogates are representations of living humans, commonly adopted to better understand human response to impacts. Though surrogates have been widely used in automotive, defence and medical industries with varying levels of biofidelity, their primary application in the sporting goods industry has been through primitive rigid anvils used in assessing personal protective equipment (PPE) effectiveness. In sports, absence from competition is an important severity measure and soft tissue injuries such as contusions and lacerations are serious concerns. Consequently, impact surrogates for the sporting goods industry need a more subtle description of the relevant soft tissues to assess impact severity and mitigation accurately to indicate the likelihood of injury. The fundamental aim for this research study was to establish a method to enable the development of superior, complementary, increasingly complex synthetic and computational impact surrogates for improved assessment of sports personal protective equipment. With a particular focus on the thigh segment, research was conducted to evaluate incremental increases in surrogate complexity. Throughout this study, empirical assessment of synthetic surrogates and computational evaluation using finite element (FE) models were employed to further knowledge on design features influencing soft tissue surrogates in a cost and time efficient manner. To develop a more representative human impact surrogate, the tissue structures considered, geometries and materials were identified as key components influencing the mechanical response of surrogates. As a design tool, FE models were used to evaluate the changes in impact response elicited with different soft tissue layer configurations. The study showed the importance of skin, adipose, muscle and bone tissue structures and indicated up to 15.4% difference in maximum soft tissue displacement caused by failure to represent the skin layer. FE models were further used in this capacity in a shape evaluation study from which it was determined that a full-scale anatomically contoured thigh was necessary to show the full diversity of impact response phenomena exhibited. This was particularly pertinent in PPE evaluations where simple surrogate shapes significantly underestimated the magnitudes of displacements exhibited (up to 155% difference) when rigid shell PPE was simulated under impact conditions. Synthetic PDMS silicone simulants were then fabricated for each of the organic soft tissues to match their dynamic responses. The developed simulants exhibited a superior representation of the tissues when compared to previous single material soft tissue simulant, Silastic 3483, which showed 324%, 11,140% and -15.8% greater differences than the PDMS when compared to previously reported target organic tissue datasets for relaxed muscle, skin and adipose tissues respectively. The impact response of these PDMS surrogates were compared in FE models with previously used single material simulants in representative knee and cricket ball sports impact events. The models were each validated through experimental tests and the PDMS simulants were shown to exhibit significantly closer responses to organic tissue predictions across all impact conditions and evaluation metrics considered. An anatomically contoured synthetic thigh surrogate was fabricated using the PDMS soft tissue simulants through a novel multi-stage moulding process. The surrogate was experimentally tested under representative sports impact conditions and showed a good comparison with FE model predictions with a maximum difference in impactor displacements and peak accelerations of +6.86% and +12.5% respectively at velocities between 2 - 4 m.s-1. The value of increased biofidelity in the anatomical synthetic and virtual surrogate thighs has been proven through the incremental adoption of important surrogate elements (tissue structures, material and geometries). The predictive capabilities of each surrogate have been demonstrated through their parallel developments and staged comparisons with idealised organic tissue responses. This increase in biofidelity is introduced at modestly higher cost compared to Silastic 3483, but, given the benefits of a more representative human impact response for PPE evaluations, this is shown to be worthwhile.

Le but de cette étude est de concevoir, fabriquer et caractériser une puce microfluidique afin de mettre en oeuve la capture de nanoparticules magnétiques fonctionnalisées en vue de la reconnaissance d’anticorps spécifiques (couplage d’une très grande spécificité et sensibilité). Après avoir modélisé et simulé les performances de la microbobine intégrée dans le canal de la puce microfluidique en prenant soin de limiter la température du fluide à 37°C, la capture devant être effective, le microsystème est fabriqué en salle blanche en utilisant des procédés de fabrication collective. La fabrication du microdispositif en PDMS a aussi donné lieu à l’optimisation de procédés de modification de surface afin d’assurer la ré-utilisation du microdispositif (packaging réversible) et la limitation de l’adsorption non spécifique. L’immobilisation des anticorps su les billes (300 nm) a été menée à l’intérieur du canal en utilisant un protocole de type ELISA éprouvé. Le procédé a montré qu’il était également efficient pour cet environnement puisque nous avons pu mettre ne évidence la capture de nanoparticules / In this study, a concept of microfluidic chip with embedded planar coils is designed and fabricated for the aim of trapping effectively functionalized magnetic nanobeads and immobilizing antibody (IgG type). The planar coils as a heart of microfluidic chip is designed with criterion parameters which are optimized from simulation parameters of the maximum magnetic field, low power consumption and high power efficiency by FE method. The characterization of microcoils such as effectively nanobeads (300 nm) at low temperature (<37oC) is performed and confirmed. The channel network in PDMS material is designed for matching with entire process (including mixing and trapping beads) in microfluidic chip. A process of PDMS’s surface modification is also carried out in the assemble step of chip in order to limit the non-specific adsorption of many bio substances on PDMS surface. The microfluidic chip assemble is performed by using some developed techniques of reversible packaging PDMS microfluidic chip (such as stamping technique, using non-adhesive layer, oxygen plasma combining with solvent treatment). These packaging methods are important to reused microchip (specially the bottom substrate) in many times. The immobilization of antibody IgG-type is performed inside microfluidic chip following the standard protocol of bead-based ELISA in micro test tube. The result showed that IgG antibodies are well grafted on the surface of carboxyl-beads (comparing to result of standard protocol); these grafted antibodies are confirmed by coupling them with labeled second antibody (Fab-FITC conjugation).

The need for safe and functional implants has led to anincreased demand for improved biomaterials. The performance invivo depends on the interaction between the biologicalsurrounding and the surface of the material. By tailoring thesurface of a material with suitable bulk properties,biomaterials with an ability to interact with the biologicalsystem in a specific and controlled way are obtained. Siliconeelastomers have been used as biomaterials for several decades,but it is widely recognized that they are difficult to modifyby the conventional methods used for organic polymers due tothe partly inorganic structure of silicone. This thesis presents a strategy to obtain siliconebiomaterials by covalent coupling of molecules to the surfaceusing silicon chemistry. The first step is to introduce Si-Hgroups onto the surface of silicone elastomers by plasmatreatment. The second step is to react a terminal double bondof a molecule with the formed Si-H group by a catalyzedhydrosilylation reaction. The coupled molecule may eitherprovide the desired properties itself, or have a functionalitythat is able to couple another molecule with suitablecharacteristics. The influence of plasma treatment in hydrogen, argon andoxygen on the silicone elastomer was characterized by X-rayphotoelectron spectroscopy (XPS). To quantify the effect ofplasma treatment, the method of ternary XPS diagrams wasdeveloped. It was found that undesired silica-like layers wereformed under severe treatment conditions. Argon plasma at lowpower and short treatment time was the most suitable parametersetting. Subsequent hydrosilylation grafting ofallyltetrafluoroethylether, aminopropylvinylether andN-vinylformamide showed that it was possible to functionalizethe surface via a covalent link to the surface. The primaryamino groups introduced onto the surface were accessible forfurther coupling reactions. Heparin surfaces were obtained by acoupling reaction with the introduced amino groups. Keywords:Silicone elastomers, PDMS, XPS, ESCA, surfacemodification, plasma

The field of micro total analysis systems (μTAS) aims at developments toward miniaturized and fully integrated lab-on-a-chip systems for applications, such as drug screening, drug delivery, cellular assays, protein analysis, genomic analysis and handheld point-of-care diagnostics. Such systems offer to dramatically reduce liquid sample and reagent quantities, increase sensitivity as well as speed of analysis and facilitate portable systems via the integration of components such as pumps, valves, mixers, separation units, reactors and detectors. Precise microfluidic control for such systems has long been considered one of the most difficult technical barriers due to integration of on-chip fluidic handling components and complicated off-chip liquid control as well as fluidic interconnections. Actuation principles and materials with the advantages of low cost, easy fabrication, easy integration, high reliability, and compact size are required to promote the development of such systems. Within this thesis, liquid displacement in microfluidic applications, by means of expandable microspheres, is presented as an innovative approach addressing some of the previously mentioned issues. Furthermore, these expandable microspheres are embedded into a PDMS matrix, which composes a novel thermally responsive silicone elastomer composite actuator for liquid handling. Due to the merits of PDMS and expandable microspheres, the composite actuator's main characteristic to expand irreversibly upon generated heat makes it possible to locally alter its surface topography. The composite actuator concept, along with a novel adhesive PDMS bonding technique, is used to design and fabricate liquid handling components such as pumps and valves, which operate at work-ranges from nanoliters to microliters. The integration of several such microfluidic components promotes the development of disposable lab-on-a-chip platforms for precise sample volume control addressing, e.g. active dosing, transportation, merging and mixing of nanoliter liquid volumes. Moreover, microfluidic pumps based on the composite actuator have been incorporated with sharp and hollow microneedles to realize a microneedle-based transdermal patch which exhibits on-board liquid storage and active dispensing functionality. Such a system represents a first step toward painless, minimally invasive and transdermal administration of macromolecular drugs such as insulin or vaccines. The presented on-chip liquid handling concept does not require external actuators for pumping and valving, uses low-cost materials and wafer-level processes only, is highly integrable and potentially enables controlled and cost-effective transdermal microfluidic applications, as well as large-scale integrated fluidic networks for point-of care diagnostics, disposable biochips or lab-on-a-chip applications. This thesis discusses several design concepts for a large variety of microfluidic components, which are promoted by the use of the novel composite actuator. Results on the successful fabrication and evaluation of prototype devices are reported herein along with comprehensive process parameters on a novel full-wafer adhesive bonding technique for the fabrication of PDMS based microfluidic devices. / QC 20100817