This thesis explores the use of reset control to improve disturbance rejection in systems subjected to wideband noise. In the high-tech industry, there is a continuous push for improved performance, particularly in achieving fast and stable motion with nanometer-level precision. Linear controllers, though industry-standard, are constrained by limitations such as Bode's gain-phase relation. Reset control offers a promising alternative, but its behaviour under wideband disturbances remains unexplored. However, such excitation is common in the context of active vibration isolation. The wide-spectrum excitation may result in dominant slow or excessive resetting, depending on the system. In both cases, the behaviour of the reset element will be different than designed. Also, the traditional method of Describing Function (DF) will not sufficiently represent the reset element's behaviour under such conditions, as it is defined for single sinusoidal excitation. This thesis studies the Power Spectral Density (PSD) of the reset triggering signal of the reset element to understand the resetting behaviour. The method of Best Linear Approximation (BLA) is applied to enable the analysis of the reset element in frequency domain. Furthermore, a novel method is developed to shape the reset triggering signal, to achieve the desired behaviour of the reset element at the frequency of interest. Both simulations and experiments showed the feasibility of the method in achieving the desired behaviour of the reset element. Depending on the properties of the system, this will result in improvement of the disturbance rejection.

With the increasing demand for extended service life and increased precision and accuracy of precision mechanics across various industries, machine tool manufacturers face the challenge of increasing the performance of their machines. To achieve this goal, the performance of the machines must be evaluated, points of improvement must be identified and subsequently be acted on.

This thesis is focused on the evaluation of the dynamic performance of a two-axis CNC lathe, subject to vibration modes excited by movement of the machine axes. To this end, two experiment setups are designed to measure displacement at the cutting tool in either axial or lateral directions and compare these displacements against the positions and controller setpoints of the corresponding machine axes. Additionally, a finite element model is created and expanded such that it emulates the machine encoders and additional cutting tool sensors, as well as the third order motion profile that is used to excite the system. Methods for automatically processing the data from these experiments and simulations are developed in parallel, allowing for large sets of data to be processed with little effort.

These experiments show that with a maximum amplitude of approximately 0.3 μm the vibrations that occur at the tool, relative to the axis encoder, are most significant in axial direction. A qualitative comparison with the finite element model indicates a mechanical issue relating to the Z axis drive mechanism, causing additional vibrations in the system. Finally, the experiment setups, corresponding data processing methods and finite element models are reflected on and incorporated into a framework for experimental machine performance evaluation. By implementing this framework, machine tool manufacturers can further evaluate and improve the dynamic performance of their machines.

This work aims to achieve a softening moment-angle response by utilizing compliant parts. The approach uses the principle of contact release, which involves initially prestressed torsional bars arranged in series, separated by rigid bodies. By stepwise activation of the torsional bars, a softening behavior is achieved. Mechanical stops are employed to maintain the initial prestress.
A Pseudo Rigid Body Model (PRBM) is developed to calculate the moment-angle behavior of the complete prototype. The optimization of the PRBM is performed using a cost function based on the least square error. The optimization parameters are the stiffness of the torsional bars and their initial prestress. Additionally, a Finite Element Analysis (FEA) is conducted to analyze the behavior of an individual I-profile torsional bar. Experimental validation of both the PRBM and the FEA models is carried out using a prototype. The prototype is constructed based on the results obtained from the PRBM optimization to follow the desired graph closely. The results from both FEA and PRBM align closely with the predicted curves, confirming the effectiveness of the proposed approach.

The exploration of compliant mechanisms, especially those providing linear motion, has become a central focus in recent engineering research. These mechanisms, characterized by their monolithic bodies and reduced component interactions, present an attractive alternative to traditional rigid mechanisms, having reduced friction, decreased energy losses, and cost efficiencies. Despite these advantages, designing and analyzing compliant mechanisms remain difficult tasks, with particular challenges arising when aiming for large-range linear motion with high support stiffness and minimal parasitic displacement. This study presents a unique design: the combination of a flexure-based linear guide with the features of a bistable compliant switch mechanism. Because a combination of these mechanisms has not been looked into while still offering a long stroke, high support stiffness and minimal parasitic motion. Through Finite Element Analysis (FEA) and subsequent iterations, an optimized model was conceived, prioritizing stiffness and minimizing undesired motions. This optimized design was turned into a 3D printed prototype. The experiments with the prototype confirmed the results of our computational insights, showcasing the parasitic motion of the prototype to be within 1mm, and its minimum stiffness at least being 10kN/m while still having a stroke of 70mm. A case study has been set up, of which the proof of concept works and is validated.

Conventional hydrostatic bearings are not suited to use on surfaces with changing curvature which limits their applicability. To improve on this, deformable hydrostatic bearings are currently being researched. The goal of this work is to develop hydrostatic bearings with a single fluid supply that work for sinusoidal counter surfaces.

The bearing pad of a deformable hydrostatic bearing must be able to conform to the shape of the sinusoidal counter surface at any position while being able to carry high loads. Previous research has also pointed out that a deformable hydrostatic bearing should have a triangular contact pressure profile for compression against a flat surface. This way a bearing without preference towards either the concave or convex part on a sinusoidal counter surface should be obtained.

The design problem is translated into a problem formulation for the synthesis of compliant mechanisms with topology optimization. In this formulation strain energies of the design domain under different load cases are utilized to achieve selective stiffness. In an extension of the formulation, the reaction forces under uniform vertical compression of the structure are required to attain a triangular shaped profile. This requirement is relaxed in the second extension where the reaction forces are only required to monotonically decrease from the middle to the sides.

Optimizations without requirement on the reaction profile yield interesting structures with clear-cut boundaries. When implementing the requirement on the reaction profile, there appears to be a trade-off between achieving the prescribed shape and the presence of grey area and artifacts in the designs. The optimizer tends to employ distributed compliance to meet the requirement, which can compromise the load-bearing capabilities of the structures.

Evaluation of selected results with a contact model predicts that numerous obtained designs could function as a hydrostatic bearing on a sinusoidal counter surface with an amplitude of 11.3 mm. Additional effort with a more realistic evaluation model is required to draw a definite conclusion on the performance.

In the semiconductor industry, the demand for a continuously higher throughput puts tough requirements on the force capacity and accuracy of actuators. Reluctance actuators, as a promising alternative to the present state-of-the-art Lorentz actuators, are capable of achieving much higher force densities. On the other hand, reluctance actuators suffer from strong intrinsic nonlinearity, such as the quadratic and position-dependent current-force relation, negative stiffness, and magnetic hysteresis, making accurate control challenging. These nonlinear effects can be much suppressed if, instead of controlling the current through the driving coil, the magnetic flux in the actuator core is controlled. As a preparation step for implementing the flux control experimentally on reluctance actuators, this project aims to design and realise the necessary hardware for flux control implementation, including a reluctance actuator prototype and a test setup, and then dynamically calibrate the actuator for flux measurement and control. More specifically, a hybrid reluctance actuator prototype that is suitable for flux control, as well as a test setup for measuring the position and dynamic force output of the actuator, are designed and realised. Using this setup, relations between key variables of the actuator are measured to enable nonlinear compensation in the current and flux control, and a hybrid flux measuring scheme using a Hall sensor and a sense coil is proposed and experimentally implemented. Although noise in the flux measurement is effectively minimised, the result shows a mismatch in the frequency domain between the outputs of the Hall sensor and the sense coil, where further investigation is needed to improve the feasibility of this approach. Furthermore, a hybrid force measuring scheme is proposed and implemented where errors of the load cell due to high-frequency dynamics of the setup can be compensated using acceleration measurements, leading to a wider frequency range for the force measurement. With the realised hardware and the obtained measurements from this project, the next step in future works would be to experimentally implement the flux control on a reluctance actuator and compare its performance with the standard current control.

Twisted and coiled polymer muscles (TCPMs) are a type of artificial muscle with a remarkable power-to-weight ratio. However, actuation dynamics are slow compared to other artificial muscles. This work aims to improve dynamic performance by incorporating redundancy. Specifically, this work examines if TCPM bundles of heterogeneous geometries containing high-force low-bandwidth actuators and low-force high-bandwidth actuators have a substantially better tracking performance than that of bundles of homogeneous geometries. First, a white-box model was created to simulate TCPM dynamics as a function of geometric parameters. The model revealed fiber diameter is the only geometric parameter that represents a trade-off between TCPM bandwidth and maximum realizable force for isometric force tracking. Next, an optimum feedforward controller was designed to distribute the reference among redundant actuators. Finally, a brute-force optimization was conducted to find the optimum configurations of heterogeneous and homogeneous TCPM bundles and the associated tracking performances. Optimal homogeneous configurations outperformed all heterogeneous configurations irrespective of number of TCPMs in parallel or reference signal. For unidirectional configurations, a nontrivial fiber diameter optimizes tracking performance. For antagonistic configurations, tracking performance improves monotonically with increasing fiber diameter.

Capillary self-alignment shows the potential to keep small components aligned using soft liquid contact when disturbed by external forces. Whilst this concept has been promising, mostly static situations have been investigated, making it unclear how a receptor-droplet-chip (RDC) configuration behaves dynamically and where its stability limits are.

The primary aim of this research was to determine the stability limits of such a configuration and explore its failure modes. Two non-linear analytical models simulate the situation and investigate the parameters exposed to the system. These models predict the relative lateral displacement of the chip and its tilt, which can predict when failure should occur. Results from Surface Evolver, a computational software, and experimental data from a laboratory setup validate these models. A shaker induced one DOF linear accelerations, and high-speed cameras observed these tests. The validated model was then used to investigate the effects of variables such as lubricant height, chip dimensions and acceleration magnitudes.

This project gives an insight into how a chip behaves dynamically on a liquid meniscus and was the first to investigate the tilt of the component during accelerations. The key finding is a stable region showing what amount of liquid results in reaching the highest possible acceleration without failure. Secondly, the models and experiments showed that smaller liquid volumes and chip sizes increase the attainable acceleration. Chips of L = 2mm were able to be accelerated up to a = 27g without failure, parts of L = 1mm even reached an acceleration of a = 35g at which the RDC configuration remained intact. Moreover, possible eigenfrequencies were observed, and the system could re-align at specific frequencies from a tilted pose. Therefore, this research has demonstrated that, under the right conditions, capillary interfaces can be useful to transport components under high accelerations.

Mobile electronics or remote devices like sensors for the Internet of Things (IoT) require energy storage with increasing energy densities, approaching the limits of the widespread lithium batteries. At the same time, lithium batteries are subject to growing criticism about their environmental impact due to their raw materials and their mining methods. Chemical energy seems to be a promising alternative and the development of micro combustion to make use of it is already for almost 30 years the topic of ongoing research. Recent studies show that compliant combustion engines are a good solution to mitigate some problems like leakage and friction and thus, achieve higher efficiencies. However, the most prevalent problem is heat losses through the walls because of the high surface-to-volume ratio in small-scale combustion. In this work, a design concept was proposed that not only insulates the compliant combustion chamber of the actuator for Flapping Wing Micro Air Vehicles (FWMAV) but also uses the waste heat to generate electricity. That was done by attaching Thermoelectric (TE) modules to a support structure and thermally connecting them with porous media to the oscillating combustor wall. The actuator uses the heat and pressure generated by the catalytic reaction of hydrogen peroxide and therewith, only emits steam and oxygen. With the help of a one-dimensional steady-state heat resistance model and a one-dimensional transient heat resistance-capacitance model, an optimum design was found. With that, it is possible to create a temperature difference sufficient enough to enable energy harvesting by TE modules. That was achieved by implementing an exhaust gas recirculation which
transfers heat to the modules and simultaneously serves as a heating blanket for the combustor. A downside of the design is the relatively long time it takes to reach its steady state and especially for FWMAV, the increase in weight.

Hexapod robots are becoming increasingly popular for navigating and exploring irregular and inhospitable environments. The Lunar Zebro is such a hexapod robot that has been developed to be the lightest and smallest rover to explore the moon. The purpose of the Lunar Zebro is to navigate the surface of the moon where it will encounter difficult terrains such as regolith (Lunar Soil) and rocks. To date, not much focus has been placed on the design of the Lunar Zebro legs to increase the rover’s trafficability on these terrains. Furthermore, the designs of rover wheels are constantly changing, and new lessons are being learnt from each new mission that the guidelines for designing the wheels and legs of rovers are constantly evolving. Using a Biomimicry design framework, the legs of the Lunar Zebro were redesigned using strategies adopted by animals in nature to improve the trafficability of the legs on granular and rocky terrains. Several concepts were created and narrowed down to four final designs (Hair, Web, Paddle and Claw Legs) that were further developed and manufactured. In addition to this, the original leg design and a flat leg with no grousers were developed for testing.

Several different single-leg tests were performed on the legs to measure their sinkage, thrust, rolling resistance and drawbar pull on regolith and rocky surfaces. The results showed that hair has the potential to reduce sinkage on regolith. However, the results from the single-leg tests on regolith also indicated that there is not much difference in the drawbar pull performance of the different legs no matter what features were included on the bottom of the leg or if the leg had grousers or not. In contrast, the results of the experiments performed on rocky surfaces showed that legs with claw-like grousers performed the best.

From there a hybrid leg was designed using the finding of the single-leg tests and this leg was tested on a basic Lunar Zebro prototype against the original Lunar Zebro leg design to determine if the hybrid improves the trafficability of the rover on both regolith and rocky terrains. These tests included testing the drawbar pull of the rover at 100 % slip and measuring the distance travelled by the rover over five steps. The results showed that on regolith terrain, the design of the leg does not have a big impact on the performance of the rover. However, the results showed that on rocky terrain the rover with hybrid legs travelled roughly 10 % further over five steps and had a drawbar pull roughly 10 % larger than the rover with original legs.

This thesis creates and validate a model for predicting the static stiffness of a linear crossed roller bearing as a function of its key design variables within a range of 20%. Linear crossed roller bearings play an essential role in the high-tech industry, this kind of bearing is used for precision applications such as microscopes, optical systems, and semiconductor production equipment. The performance of all these systems is related to the ability to produce and control a certain motion of a body. Bearings facilitate this motion, they are designed such that the resistance in the direction of movement is minimal. Generally, in all other directions, the bearing should be as stiff as possible as a high bearing stiffness is beneficial for controlling the motion. Stiffness is beneficial as it contributes to achieving a high control bandwidth and minimizes sensitivity to external forces or vibrations. Mechatronic systems have to meet very strict targets in terms of positioning accuracy and settling times. The dynamics of mechatronic systems is dominantly dependent on the performance of their bearings. Nowadays, the performance of a machine is predicted before building the physical machine through virtual prototyping. Virtual prototyping is becoming increasingly important: an accurate model in an early stage of a development project minimizes the risk of not achieving specifications in a later stage. For accurate modelling of precision machines using linear cross roller bearings, reliable data on the stiffness characteristics of these bearings is required. This thesis performs a step-by-step validation of models. Starting with an empiric determination of the load-stiffness relation for an individual cylindrical roller-rail contact as this is not affected by the uncertainties present in bearing assemblies. This is followed by an analysis of the tangential effects that come into play when a roller is loaded at 45°. Once the roller-rail model at 45° is validated, this empirical roller-rail model is used to develop a model that predicts the stiffness of linear roller bearing assemblies.

Roll-to-Plate (R2P) nanoimprint lithography (NIL) is a fabrication process with the potential for low-cost high-volume fabrication of micro- and nano-patterns on large areas. As a result, R2P is regarded as a promising production process for micro- and nano-scale features which exceed the conventional wafer sizes. In addition, the process has the potential to scale up the production of wafer based textures at a low costs, which would make the implementation of nano-features in products more accessible. To compete with alternative fabrication techniques, R2P NIL must meet requirements set by the industry for position accuracy and system repeatability. In this thesis, a study is performed on the positioning of micron- and nano-sized features in a two-dimensional plane in collaboration with Morphotonics, a company based in Veldhoven the Netherlands. The metrology of overlay from the semi-conductor industry is applied to NIL, to quantify the relative displacement and deformations of a given nano-pattern. Experimental results, obtained in a preliminary phase of the project, characterized the current overlay performance of machines available at Morphotonics. From the test results, a positioning error of the complete nano-pattern was found, in combination with a complex deformation which affected the relative distance of the imprinted features. Based on the characterization of overlay in R2P NIL, an interest in deformations of the flexible mould is developed. Concepts from roll-to-roll (R2R) production processes are used, to quantify the deformations observed in experiments. In R2R processes, deformation of a thin elastic material (web) is quantified by applying a beam model to the web. Using the beam model, the deformation of a web segment, constrained by two rollers, could be quantified in both quasi-static and dynamic conditions. The models based on literature of lateral web dynamics, combined with experimental results obtained at Morphotonics, have given insight in the behaviour of the elastic mould during the imprint process. The models describe the transverse motion of the web on a roller surface and the deformation of the web in between two consecutive rollers in a dynamic system. Verification of the model was performed on the machines of Morphotonics. The transverse motion of the elastic mould over the rollers was clearly observed in the measurement, including a repeatable deformation pattern of the imprints as a result of a forced disturbance exposed to the system. Both of which, were expected to be observed based on the modelled expectations. Nevertheless, the exact error magnitude did not meet the expectations and only a rough relation between model and experimental results was found. In spite of the verification results not perfectly matching the model, the theory applied in this thesis has given insight in the behaviour of elastic materials in roller based transport. From the observations and gathered knowledge of this thesis, the performance of overlay in large scale roll-to-plate nanoimprint lithography has been characterized and considerations for future imprint modules have been made.

In recent developments in the field of multi-material additive manufacturing, differences in material properties are exploited to create printed shape memory structures, which are referred to as 4D-printed structures. New printing techniques allow for the introduction of prestresses in the specimen during manufacturing. This research focuses on bi-polymer 4D-printed structures, where the transformation process is based on a heat-induced stiffness transition in one of the materials and an initial prestress in the other material. Upon the decrease in stiffness, the prestress is released, causing a bending behaviour. A methodology to find the design of 4D-printed structures is developed, where a finite element model is combined with a density-based topology optimization method to describe the material layout. This modeling approach is verified by a convergence analysis and validated by comparing its numerical results to analytical and published data. The use of topology optimization to design 4D-printed structures is explored by applying the methodology to a variety of design problems. Bi-layer designs are printed with acrylonitrile butadiene styrene (ABS) and thermoplastic polyurethane (TPU), of which the latter material is prestressed during the Fused Filament Fabrication (FFF) printing process. Tests are performed with these printed samples, with the goal to find the prestress value and to further validate the modeling approach by comparing the numerical results to the experimental results. Once the prestress value is found, topology optimized samples are printed and tested to evaluate their performance.

The computational modeling of crack propagation is important in aerospace and automotive industries because cracks play a significant role in structural failure. In comparison with the quasi-static case, inertial effects are significant in dynamic crack propagation. The classical numerical tool for modeling dynamic crack propagation is the standard finite element method (FEM). However, the method suffers from problem of mesh-dependency. Enriched finite element procedures such as the eXtended/Generalized finite element method (X/GFEM) are able to solve this issue by adding enriched degrees of freedom (DOFs) to nodes of the original mesh, thereby providing full decoupling between the mesh and cracks. Yet, X/GFEM suffers from other issues that are inherent to the formulation such as the need for more complicated methods for applying non-zero Dirichlet boundary condition, and an intricate computer implementation. In the thesis, the Discontinuity-Enriched Finite Element Method (DE-FEM) is used to simulate dynamic crack propagation as an alternative to X/GFEM. DE-FEM also decouples cracks from the background mesh while solving X/GFEM’s aforementioned issues. This is achieved by adding enriched DOFs along cracks. Prescribing non-zero essential boundary conditions is as straightforward as in standard FEM. Moreover, DE-FEM’s enrichment strategy provides simplicity on crack propagation, as the crack tip node can be transformed into a crack segment node directly with little effort. Also, fewer enriched DOFs than in X/GFEM are appended while advancing cracks. In the numerical examples of either stationary crack with dynamic loading or dynamically propagating cracks, DE-FEM is able to reproduce analytical and/or experimental results. Both Bathe’s time integration method and the classical Newmark constant average acceleration method are applied and compared. Results show DE-FEM is a suitable candidate to solve dynamic fracture problems.

Energy harvesting from renewable energy sources has become more popular in the last decades than ever before. New energy sources consisting of human input vibrations are hopeful alternatives for powering wearable low power electronics. These sensors currently rely on the lifetime of the battery and come along with high maintenance costs in case of a replacement. Many designs found within literature are based on the working principles of linear resonant energy harvesters, consisting of high output generation when they are exited on their resonance frequency. However, if the vibration energy harvester is not accurately tuned to the input signal of the real world, poor output performance can be expected. Bistability, consisting of a unique double well potential energy curve is an interesting alternative. The oscillation between the two potential wells contribute to higher output performance in comparison with resonant configurations. However, these potential wells are segregated by an energy barrier and the motion between the two wells will not always occur during excitation. A solution is found to reduce the energy barrier by means of mechanical end-stops. A mechanical model based upon beam theory is created in ANSYS and their stiffness characteristics are used as an input parameter for the dynamical model. To confirm this model a prototype is constructed and investigated. A mechanical analysis is carried out using a quasistatic forcedeflection measurement and the dynamical analysis is performed on a linear air bearing stage, consisting of a maximum stroke of half a meter being able to reproduce low frequency input excitations. It could be observed that participation of the desired trajectory between the two potential wells is enhanced and occur at lower input accelerations, as the oscillators motion is confined by means of hard mechanical end-stops. Therefore, the integration of mechanical end-stops as a design parameter for bistable energy harvesters can be considered as a viable solution to capture the kinetic energy induced by human input motion with the use of bistable mechanisms.

In the field of vibration energy harvesting, vibration energy is tranduced to electrical energy to power small, low powered devices. Many energy harvesters are produced in the form of a resonator, where resonant amplification enables efficient operation of the energy harvester. At low frequencies below 10 hz, input motions quickly increase for a constant input acceleration and resonant amplification results in energy harvesters that are too large to implement them. A solution is sought in creating a nonresonant energy harvester. The stiffness of a strongly coupled piezoelectric beam is compensated by adding negative stiffness to bring it to a near statically balanced state. This negative stiffness is embodied by attracting magnets. To model the dynamics and voltage output of the harvester, a modal analysis based distributed parameter model is used and further developed by including negative stiffness and force-displacement measurements of the stiffness compensated piezo. To investigate the mechanical behaviour of a compensated piezo, force-displacement measurements are carried out at different deformation speeds and load resistances. From these measurements, it has been observed that the stiffness of the compensated piezo strongly depends on the connected load resistance and the deformation speed. Furthermore, memory effects in piezoelectric hysteresis found in actuators such as curve alignment and wipeout have also been confirmed in force-displacement measurements. The performance of the harvester has been evaluated by exciting it on a linear air bearing stage. It has been found that for excitations between 2 and 6 hz, the error in RMS power between simulation and measurement remains below 10%. For its range of excitation, this harvester has been observed to be the most efficient with respect to prior art from literature. Therefore, stiffness compensation of a piezoelectric energy harvester can be considered as a successful method to improve the efficiency at low frequency excitation.

Bistable vibration energy harvesters are an interesting alternative to their linear counterparts. They allow for large amplitude oscillations between their stable equilibria, from which much energy can be generated. However, the stable equilibria are separated by a potential energy barrier that has to be overcome. Therefore, we cannot guarantee these oscillations, and the performance advantage diminishes. As a solution to this, a novel method of stiffness compensation in compliant bistable mechanisms is explored to lower the potential barrier. This method makes use of interaction between the buckling modes. Whereas this phenomenon is most undesired in structures due to their increasing proneness to catastrophic failure, we cleverly use it to our advantage. During the deflection required for the large amplitude oscillations, a transition between these buckling modes occurs, causing the increase in potential energy. By bringing the corresponding buckling loads closer together, the transition is eased and the potential barrier is lowered. An analytical framework was set up as a fundamental test of this method. Using a discrete analytical model of a bistable buckled four-bar linkage with torsion springs, it was shown that the potential barrier can be flattened upon matching the first two critical buckling loads, resulting in static balancing. This was achieved by making two torsion springs three times stiffer with respect to the other two springs. To put theory into practice, three compliant mechanisms were designed using the ratio between the first two buckling loads. Their force-deflection characteristics were experimentally determined and it was shown that the stiffness may be tuned according to the ratio between the buckling loads. Furthermore, it was shown that in designs having the first two buckling loads equal to each other, near zero stiffness is achieved. Hence, this method is proven a successful addition to the arsenal of methods in stiffness compensation and static balancing of compliant mechanisms.

A LED induced fluorescent technique based test setup is introduced , which is used to measure the water film thickness of compliant hydrostatic bearings. Fluorescent particles have the characteristic of absorbing light of a certain wavelength and emitting light of a higher wavelength. A LED with a wavelength of 540 nm (green light) is used to excite the Rhodamine B fluorescent particles, which have a excitation wavelength of 545 nm. The emission wavelength of the particles is 570 nm (red light) and this colour shift is captured by a RGB colour camera where the red and green channels have been used as an optical filter. The increase of intensity due to the excitation of the thin film can be coupled to the thickness of the film. The performance of the technique is validated by comparing the film height to a laser sensor measured film height for a rigid bearing containing features with known dimensions. A film thickness measurement technique is presented that is able to measure 15-1400 $\mu$m and an average accuracy of 6.4\% is achieved.

The demand for nanoparticles in various applications is increasing. These applications can only be realised if nanoparticles can be specifically arranged and patterned on a substrate. This thesis considers a direct write method that uses an aerodynamic focusing nozzle to deposit these particles. A major challenge arises when the nanoparticles are decreased in size. This research studies the control of the deposition resolution of small sized nanoparticles (≤10 nm) in an aerosol flow. A FEM model was developed to describe the particle’s path. These models were made for a converging and converging sheath gas nozzle. Due to the small particle sizes considered, it was found that only the Stokes (drag) force needs to be accounted for during modelling. The effect of different nozzle exit throat, working distance, angle and flow rate configurations are studied. Nozzle designs were evaluated using three performance criteria, namely the contraction factor, focusing ratio and line width. These describe the contraction of the particles within the nozzle system, the focusing after the nozzle system and the width of the line, respectively.
It was found that smaller angles, longer converging sections and higher velocities resulted in smaller line widths. Also, the contraction factor hardly depends on the particle size. Smaller nozzle exit throats have significantly higher focusing ratios for particles smaller than 10 nm. The working distance does not effect the contraction factor directly, indicating capabilities of deposition on non-flat surfaces. The converging nozzles can only deposit ≤10 nm particles if the nozzle exits are sufficiently small. This can cause clogging. The focusing ratios, in these nozzles, never exceeds the value of one, indicating always a larger line width than the nozzle exit diameter.
In the sheath gas nozzle system, high sheath gas ratios are essential for increasing the contraction factor and particle velocity. This prevents clogging. Introducing the sheath gas earlier in the nozzle system is more effective than at the end. The best modelled contraction factor, focusing ratio and line width achieved, using 10 nm particles, with a converging nozzle are 1.0, 0.4 and 874 microns, respectively. This nozzle has a nozzle exit throat of 400 microns and an angle of 10 degrees. However, a modelled contraction factor of 9.3, focusing ratio of 3.8 and line width of 104 microns are achieved using a converging sheath gas nozzle with a nozzle throat of 400 microns and an angle of 5 degrees. Narrower line widths are expected if the throat diameters are reduced.
For the experiments, a setup was used in which nanoparticles are generated using a spark ablation method. During these experiments it was seen that the maximum flow rate is mainly dependent on the nozzle exit throat of the aerosol channel. The line widths, during experimenting, were found to be wider due to inaccuracies in the stage and the differences in settings between the model and setup.