A pneumatic droplet generator is described, which consists of a T-junction with a nozzle fit into one opening, the second opening is connected to a gas cylinder through a solenoid valve and the third connected to a length of steel tubing. The droplet generator is filled with liquid. Opening the valve for a preset time creates a pulse of alternating negative and positive pressure in the gas above the surface of the liquid. A jet of liquid issues far enough out of the nozzle that its tip becomes unstable, detaches and forms a droplet. Experiments were conducted using water/glycerin mixtures and molten metals including tin, zinc and zinc alloy. Droplet formation was photographed and the pressure variation inside the droplet generator recorded. The effect of various experimental parameters such as nozzle size, pressure pulse width, secondary gas flow pressure, liquid viscosity and temperature on droplet formation were investigated. An analytic model of incompressible liquid motion in the nozzle is used to explain the behavior of water/glycerin solutions inside the nozzle and droplet formation. The model demonstrates that the motion of the surface is out of phase with the exciting pressure oscillation. Experiments showed the oscillation of the liquid surface prior to droplet ejection and the time lag between the pressure oscillation and droplet ejection. The model predicts that maximum liquid velocity is attained at an intermediate value of viscosity, and experiments confirmed that the largest liquid motion was achieved with this intermediate value, which eventually leads to droplet formation. Similarly, with molten metals, a simple analytical method was used to estimate the diameter of droplets. The size of tin droplets measured from experiments was in good agreement with that obtained from the model.

A pneumatic droplet generator is described, which consists of a T-junction with a nozzle fit into one opening, the second opening is connected to a gas cylinder through a solenoid valve and the third connected to a length of steel tubing. The droplet generator is filled with liquid. Opening the valve for a preset time creates a pulse of alternating negative and positive pressure in the gas above the surface of the liquid. A jet of liquid issues far enough out of the nozzle that its tip becomes unstable, detaches and forms a droplet. Experiments were conducted using water/glycerin mixtures and molten metals including tin, zinc and zinc alloy. Droplet formation was photographed and the pressure variation inside the droplet generator recorded. The effect of various experimental parameters such as nozzle size, pressure pulse width, secondary gas flow pressure, liquid viscosity and temperature on droplet formation were investigated. An analytic model of incompressible liquid motion in the nozzle is used to explain the behavior of water/glycerin solutions inside the nozzle and droplet formation. The model demonstrates that the motion of the surface is out of phase with the exciting pressure oscillation. Experiments showed the oscillation of the liquid surface prior to droplet ejection and the time lag between the pressure oscillation and droplet ejection. The model predicts that maximum liquid velocity is attained at an intermediate value of viscosity, and experiments confirmed that the largest liquid motion was achieved with this intermediate value, which eventually leads to droplet formation. Similarly, with molten metals, a simple analytical method was used to estimate the diameter of droplets. The size of tin droplets measured from experiments was in good agreement with that obtained from the model.

Zugl.: München, Techn. Univ., Diss. / !01044663X! ; Nr. 384

An imaging system was developed to visualize Drop-on-Demand (DOD) inkjet drop formation and impaction on substrate over drop sizes and impaction speeds of the magnitudes encountered in applications of inkjet printing. Using a pulsed laser, a low-speed charge-coupled-device (CCD) camera, and signal generators, the imaging system based on flash photography was used to capture sharp images with a temporal resolution of 200 ns and a spatial resolution of 0.81 micron/pixel. First, the dynamics of DOD drop formation was studied experimentally. The effects of the driving signal, which controls the piezoelectric transducer that produces the pressure pulse to drive the liquid from the reservoir through the orifice, have been examined along with those of liquid properties. The main stages of DOD drop formation, including ejection and stretching of liquid, pinch-off of liquid thread from the nozzle exit, contraction of liquid thread, breakup of liquid thread into primary drop and satellites, and recombination of primary drop and satellites, are analyzed. A necessary condition for the recombination of the primary drop and satellite and the limit for liquid thread length without breakup during contraction are proposed. Second, using the visualization system coupled with a motorized stage, micron-drop impaction on a smooth substrate was investigated over a regime of We and Oh typical for inkjet printing applications. The results indicate that scaling of micron-drop impaction from millimeter-drop impaction, based on dimensionless numbers (Oh, We and cos ), is valid. The predictions of maximum spreading ratio by six existing models agree well with experimental values for high-We impaction, but not for low-We and low-contact-angle impactions; however, the model of Park et al. predicts well for high- and low-We impaction due to its inclusion of spontaneous spreading dissipation. Fingering and splashing do not occur in the micron drop impaction on either dry solid substrates or a pre-existing liquid layer. The drying time of a micron drop deposited on a substrate is less than one second and increases as the contact angle of the drop on the substrate increases.

The research discussed in this dissertation was conducted to understand drop formation of inkjet printing with inks containing polymer. Solutions containing a water soluble polymer, poly ethylene oxide (PEO), with different molecular weights and polydispersities were used as inks. A flash photographic technique was used to visualize the whole process of DOD drop formation of dilute polymer solutions. The effects of driving signal, frequency and liquid properties on drop speed, drop size, breakup time and the formation of satellites were studied in detail. The addition of PEO increases the shear viscosity at all molecular weights, but the change is small for dilute solutions. However, the addition of a small amount of PEO can have a significant effect on the DOD drop formation process, increasing breakup time, decreasing primary drop speed and decreasing the number of satellites in some cases. The effects depend on both molecular weight and concentration. At lower molecular weights (14k and 35k g/mol), the effect of PEO was small when the drop formation process for the dilute solution was compared with that of a Newtonian liquid having similar shear viscosity, and the effect of PEO was small even at concentrations large enough that the solution does not fall in the dilute regime. As molecular weight is increased, the effects of PEO on DOD drop formation increase significantly, and the effects of concentration become important. These effects are explained by the fluid elasticity which increases with increasing in molecular weight and concentration. When the liquid jets out of the nozzle, the polymer chains are stretched, and thus depart from their ideal coiled state. As a result, an elastic stress develops in the liquid column and resists capillarity-driven pinch off from the nozzle and is responsible for the decrease in drop speed and longer breakup time. DOD drop formation data were shown to correlate closely with effective relaxation time, proposed by Tirtaatmadja based on Rouse-Zimm theory. When driving voltage amplitude is 44.2 V, two important parameters (breakup time and primary drop speed) in DOD drop formation for solutions containing monodispersed PEO and aqueous solutions containing mixtures of monodispersed PEO were closely predicted by correlation equations involving effective relaxation time . A mixture rule was developed to calculate the relaxation time for mixtures of monodispersed PEO. However, for polydispersed PEO, effective relaxation time was based on viscous molecular weight since the molecular weight distributions of the polydispersed PEO were unknown. When breakup time was plotted versus effective relaxation time for 1000k g/mol PEO, the data did not lie on the same line as that for the 100k and 300k g/mol PEO. This is believed to be due to the molecular weight distributions of the polydispersed PEO. When more than one species are present, viscous average molecular weight does not adequately account for the long chain species making up the polymer sample. DOD drop formation dynamics is highly affected by the actuating waveform, including the driving voltage, waveform shape, and frequency. The effects of parameters (jetting frequency, voltage amplitude and the shape of waveform) characterizing the signal were investigated. The open time and first drop problem were also studied. Research in this dissertation gives a better understanding of DOD drop formation process of polymer solutions, which may lead to improvement of inkjet printing quality for a variety of industry inks and polymer micro scale deposition and patterning in large areas.

Drop-on-demand (DOD) inkjet print-head has a major share of the market due to simplicity and feasibility of miniature system. The efficiency of droplet generation from DOD print-head is a result of several factors, include viscosity, surface tension, nozzle size, density, driving waveform (wave shape, frequency, and amplitude), etc. Key roles in the formation and behavior of liquid jets and drops combine three dimensionless groups: Reynolds number, Weber number and Ohnesorge number. These dimensionless groups provide some bounds to the “printability” of the liquid. Adequate understanding of these parameters is essential to improve the quality of droplets and provide guidelines for the process optimization. This thesis research describes the application of computational fluid dynamics (CFD) to simulate the creation and evolution process of droplet generation and transport of a highly viscous Newtonian fluid. The flow field is governed by unsteady Navier-Stokes equations. Volume of Fluid (VOF) model is used to solve this multi-phase (liquid-gas) problem.

Dropp-genereringstekniker är viktiga för industrier som läkemedelsindustrin, livsmedelsindustrin, kosmetikindustrin etc. Traditionella droppgenereringstekniker är dock begränsade i mängden av material som kan processas till droppform. Ett exempel inkjet som är en väletablerad teknik för att generera droppar med hög hastighet (1-10 kHz) och precision (10-20 μm), men kan bara stöta ut vätskor med låga viskositet, ungefär 10-100 gånger viskositeten av vattnet. Akustophoretisk utskrift motiv är att övervinna denna materialbegränsning och har framgångsrikt avkopplat dropputstötning från bläckviskositet. Metoden utnyttjar ickelinjära akustiska krafter för att skriva ut en stor mängd av material med hög kontroll, med viskositet som sträcker sig över fyra storleksordningar (0,5 mPa · s till 25 000 mPa · s). Emellertid är utstötningen baserad på bildandet av en hängande droppe, och i den aktuella prototypen begränsas materialpaletten av akustophoretisk utskrift genom sprider sig över munstycket, vilket begränsar den minsta tillåtnas ytspänningen till ungefär 60 mN / m. I detta arbete införs en munstycksbeläggningsteknik för att expandera mängden av utskrivbara material, med tillåtna ytspänningar så låga som 25 mN / m. Genom att utnyttja generera nanostrukturer med låg ytenergi på munstyckspetsen, tillverkas superavstötande beläggning. Grunden för nanostrukturerna genererades med hjälp av sot från ett paraffin-vaxljus. Ett robust tillverkningsprotokoll har etablerats, och beläggningen fysikaliska egenskaper och prestanda har karaktäriserats. Tre nya tillämpningsområden undersöktes, vilket demonstrerade noviteten hos denna nya metod. Detta arbete banar vägen för en ny uppsättning material som ska behandlas i en droppe-per droppe metodik. / Droplet generation techniques are essential for industries such as the pharmaceutical, food industry, cosmetic industry, etc. However, traditional droplet generation techniques are limited in the palette of materials that can processed in a droplet form. For example, inkjet which is a well-established technology to generate droplets of high speed (1-10 kHz) and precision (10-20 μm), but can only eject fluids with low viscosities, roughly 10-100 folds the one of water. Acoustophoretic printing aims to overcome this material limitation and have successfully decoupled droplet ejection from ink viscosity. The method harnesses nonlinear acoustic forces to print a wide range of materials on demand, spanning over four orders of magnitudes (0.5 mPa·sto 25,000 mPa·s). However, the ejection is based on the formation of a pendant drop, and in the current prototype, the material palette of acoustophoretic printing is limited by nozzle wetting, limiting the allowable minimum surface tension to about 60 mN/m. In this work, a nozzle coating technique is introduced in order to expand the material window by processing fluid with a surface tension as low as 25 mN/m. By leveraging self-assembling of nanostructures on the nozzle tip, superamphiphobic coating is successfully manufactured by using a candle soot template.A robust manufacturing protocol has been established, and the coating characterized in its physics and performance.

Trilayer polypyrrole microactuators that can operate in air have previously been developed. They consist of two outer layers ofthe electroactive polymer polypyrrole (PPy) and one inner layer of a porous poly(vinylidene flouride) (PVDF) membranecontaining a liquid electrolyte. The two outer layers of PPy are each connected with gold electrodes and separated by the porousPVDF membrane. This microtool is fabricated by bottom-up microfabrication However, porous PVDF layer is not compatible with bottom upmicrofabrication and highly swollen SPE suffers from gold electrode delamination. Hence, in this MSc project/thesis a novelmethod of flexible electrode fabrication with conducting polymers was developed by soft lithography and drop-on-demandprinting. The gold electrodes were replaced by patterned vapor phase polymerized (VPP) poly(3,4-ethylenedioxythiophene) (PEDOT)electrodes due to its high electrical conductivity and versatile process ability. The replacement of the stiff gold electrodes byflexible and stretchable PEDOT allowed high volume change of the material and motions. The PEDOT electrodes werefabricated by patterning the oxidant iron tosylate using microcontact printing and drop-on-demand printing. Moreover, thePVDF membrane has been replaced by a nitrile butadiene rubber/poly(ethylene oxide) semi-interpenetrating polymer network(IPN) to increase ion conductivity and strechability and hence actuator performance.

The objective of the research was to develop fundamental understanding of the process of deposition of complex mixtures by the inkjet method. The rheological properties and DOD drop formation dynamics of carbon black pigmented inkjet inks were investigated. It was found that the suspension microstructure responses to bulk motions, leading to shear rate and time dependent shear viscosity. However, DOD drop formation dynamics of highly pigmented inkjet ink and pure Newtonian fluid is similar even though shear rate up to 105 s-1 exists during inkjet jetting process. A proposed explanation for these observations is that the shearing time during DOD drop ejection is insufficient for changing and stabilizing the microstructure of the suspension. The effects of signal amplitude and jetting frequency on DOD drop formation dynamics of pure Newtonian fluids were investigated. A transition of DOD drop formation dynamics when the inkjet nozzle is switched from idle to jetting was identified. A qualitative investigation of DOD drop impaction and post-impaction behavior on inkjet paper and textiles was carried out. Dynamics of DOD drop accumulation and spreading on the substrates and final ink distribution show drastic differences between these two substrates.

As drop-on-demand (DOD) applications continue to gain ground in desktop inkjet-printing, 3D printing, fluid mixing, and other areas the demand for higher frequency operations are beginning to push against the current physical boundaries in DOD technology. The current research is exploring the possibility of controlling drop volume and velocity at high frequency ranges where meniscus vibrations can occur between drop formations and affect drop formation characteristics. A periodic voltage is applied to a piezoelectric disk in order to generate pressure fluctuations in a single nozzle droplet generator, causing the fluid meniscus at the nozzle to vibrate. A single stronger pulse is then superimposed over the periodic waveform at different phases in order to drive drop ejection. The characteristics of the resulting drop, specifically the volume and velocity, are experimentally measured using a high speed camera with precise timing control. The results of these experiments are then compared to a lumped element model (LEM) developed for the droplet generator geometry used. Within the LEM model framework, special attention was given to the definition of a novel method by which one can measure drop volume within an electroacoustic circuit and also allow meniscus dynamics to affect present and future drop formations. Experimental results indicate a strong dependence of both drop volume and drop velocity on the phase of the vibrating meniscus at the start of drop formation. Positive meniscus displacements and momentums resulted in large drop volumes and velocities while negative displacements could reduce drop volume or altogether eliminate drop formation. Specifically, positive displacements and momentum of a vibrating meniscus could lead to drop volumes approximately 50% larger than the original drop volume without a vibrating meniscus. Meanwhile, negative meniscus displacements and momentums were shown to have the ability to completely prevent drop formation. Additional potential for drop characteristic control with a vibrating meniscus is discussed alongside observations on the stabilizing affect the vibrating meniscus appears to have on drop velocity as a function of time. Also, flow visualization of the drop formation is provided to demonstrate the added affect the meniscus vibrations have on the drop shapes and break-off profiles. The LEM model presented demonstrates qualitative agreement with the experimental model, but fails to quantitatively predict drop volumes. Sources of error for the LEM model and potential improvements are discussed. / Graduation date: 2012