The natural instability and breakup of electrified cone jets leads to the formation of chargeddroplets with size and charge distribution of the droplets governed by the physical properties and electrospraying conditions under consideration. This thesis develops a leaky-dielectric model to study the axisymmetric breakup of an electrified jet, using the phase field method to treat interfacial phenomena. The model is used to analyze the breakup in a wide range of the Taylor number (Γ), the Ohnesorge number (Oh) and the wavenumber relevant to electrosprays operating in the cone-jet mode. The phase field technique accurately captures the behavior of the jet after pinch-off and predicts the formation of primary and satellite droplets. The numerical results are compared with existing experimental and numerical studies, extending them to account for the formation of sub-satellite droplets. It is found that for highly viscous jets, Oh ≫ 1, the number of sub-satellite droplets generated increases with the Taylor number when compared to low viscous jets, hence widening the size distribution of droplets. At fixed Γ and Oh the primary droplets are charged to an approximately constant ratio of the Rayleigh charge limit, regardless of the wave number. Furthermore, the primary droplets are charged below the Rayleigh limit for Γ ≲ 1.5, and charged above the Rayleigh limit when Γ ≳ 1.5. Thus, most primary droplets are expected to be unstable at Taylor numbers exceeding 1.5.

With a rapid increase in the interest in use of nanosatellites in the past decade, finding a precise and low-power-consuming attitude control system for these satellites has been a real challenge. In this thesis, it is intended to design and test an electrospray thruster system that could perform the attitude control of a 1-unit CubeSat.

Firstly, an experimental setup is built to calculate the conductivity of different liquids that could be used as propellants for the CubeSat. Secondly, a Time-Of-Flight experiment is performed to find out the thrust and specific impulse given by these liquids and hence selecting the optimum propellant. On the other hand, a colloidal thruster system for a 1-U CubeSat is designed in Solidworks and fabricated using Lathe and CNC Milling Machine. Afterwards, passive propellant feeding is tested in this thruster system. Finally, the electronic circuit and wireless control system necessary to remotely control the CubeSat is designed and the final testing is performed.

Among the propellants studied, Ethyl ammonium nitrate (EAN) was selected as the best propellant for the CubeSat. Theoretical design and fabrication of the thruster system was performed successfully and so was the passive propellant feeding test. The satellite was assembled for the final experiment but unfortunately the microcontroller broke down during the first test and no promising results were found out. However, after proving that one thruster works with passive feeding, it could be said that the ACS testing would have worked if we had performed vacuum compatibility tests for other components beforehand.

ABSTRACT OF THE DISSERTATION

Multiplexing of electrospray sources for space propulsion and physical sputtering

By

Enric Lluís Grustan Gutiérrez

Doctor of Philosophy in Mechanical and Aerospace Engineering

University of California, Irvine, 2015

Professor Manuel Gamero Castaño, Chair

The present work explores what can be achieved when the fields of microfabrication and electrospray atomization intersect. The electrospraying of conductive liquids is a well documented technique used heavily for mass spectrometry but also for drug encapsulation or electric propulsion among other applications.

Chapters 2 and 3 of the dissertation focus on the development of a miniaturized array of sixty-four electrospray sources, determining the design requirements and how to adapt them to microfabrication processes. The prototypes response to mass flux and emitting voltage variations is characterized by recording the currents at the emitter, extractor and collector. To determine whether the source can run continuously for prolonged periods of time, it is left operating until malfunction; the maximum life span is 12 hours.

The most advanced prototype undergoes Time-of-flight spectrometry analysis to evaluate its capabilities for satellite propulsion; with resulting mass flux ranging from 2.07·10-9 to 2.81·10-8 kg/s, a total maximum thrust of 34 µN, an Isp varying from 124 to 245s, with specific charge between 697 and 3142 C/kg and a maximum thrust efficiency of 70%.

In the final portion of the document we present the effects of bombarding Si, SiC, InAs, InP, Ge, GaAs, GaSb and GaN with high kinetic energy projectiles from a single emitter electrospray source. We investigate the structure of the damaged surface and sputtered volume by atomic force and scanning electron microscopy and stylus profilometry. The dependance of substrate nature, kinetic energy and projectile dosage on the final topography of the processed slice is discussed together with sputtering rate and yield. The maximum sputtering rates for the technologically interesting SiC and GaN are 220 and 630 nm/min respectively.

Gold, AZ4620 and Shipley 1827 photoresists layers are deposited on a silicon substrate and sputtered to find their adequacy as microfabrication masks. The sputtering rate selectivity between mask and substrate is calculated for different impact velocities; the maximum value for gold is 64.17, at low speeds Shipley and AZ selectivity is virtually infinite. Finally as proof of concept some intricate structures are carved on silicon using AZ4620.

Electrospray propulsion stands out as a unique electric propulsion technology, particularly well-suited for the growing popularity of SmallSats, spacecraft with a mass below 500 kg. Microfabrication techniques are used to create highly dense arrays of emitters allowing to scale up the thrust and process the low power available in SmallSats while maintaining a very high efficiency.However, in electrospray propulsion, overspraying of the extractor and accelerator electrodes poses a significant challenge, as the accumulation of fluid on these electrodes can lead to the formation of electrically conducting films and the catastrophic shorting of power supplies. Therefore, it is crucial to have a detailed understanding of the physics of the beam and the ability to model it accurately to design electrodes effectively and predict potential impingement during the long operational periods typical of electric thrusters, which can span several years. The model presented is used to study the composition of the beam emitted by the microfabricated electrospray thruster developed by our group. The model combines two different approaches: a Lagrangian model for the inner region near the jet breakup where the trajectories of all individual charged droplets are integrated simultaneously, and where Coulomb repulsion between droplets is fully captured while integrating the equations of motion; and an Eulerian model for the outer region where the structure of the beam is obtained by computing the envelopes of beamlets in which the population of droplets is subdivided. The two regions are coupled by the initial conditions of the envelopes, which are calculated with the solution of the inner problem; and by the influence of the space charge of each region on the electric field. This strategy greatly reduces the computational time and allows to simulate the whole beam while retaining the important particle-on- particle Coulomb interaction where is significant.

Progress on miniaturization techniques in the last two decades has made it possible to scaledown spacecraft subsystems to smaller sizes and power levels, which resulted on a new category of satellites with mass below 600 Kg, called smallsats. The scale-down of satellites into smallsats, the decreasing service price of commercial rocket launchers, and standardization in the manufacturing of smallsats have boosted space technology activity. However, many smallsats in orbit lack a propulsion system due to the complexity of scaling it down to the volume, mass, and low power requirements. Electrospray propulsion is a natural fit for the micropropulsion required by smallsats because it is a soft ionization technique that does not involve the formation of a plasma. A single electrospray emitter, electrospraying a high conductive liquid such as an ionic liquid, efficiently converts electric power into beam kinetic power while operating at the available power levels. Successful integration of electrospray emitter arrays requires, however, both microfabrication expertise and detailed knowledge of the fundamentals of electrosprays. The first part of this work integrates an in depth study of a highly conductive liquid for its interest for electrospray propulsion with a scalable microfabrication method of silicon micro-emitters for a compact microfluidic electrospray propulsion systems. Its components, that include an emitter array electrode with fractal-like microchannels etched on the back side and perpendicular to the out-of-plane emitters, an extractor electrode, and a supporting micromachined glass substrate, are permanently bonded and precisely aligned using anodic bonding. The number of emitters, the hydraulic resistance of the microfluidic system, and the gap between the extractor electrode and the emitter electrode can be tailored during the initial design and fabrication steps to achieve the desired pressure and voltage operation range. The system presented demonstrates good performance, uniformity, synchrony of emission in each emitter part of the array, and a rapid response to the applied pressure in the propellant reservoir. The deposition of counter ions during operation is identified as key performance issue and addressed by adding an interface of platinum on the emitter array. The performance tests (86 hours and ongoing) demonstrate the largest operational lifetime of a microfabricated electrospray source with capillary-like emitters actively fed. The thrust measurements show thrust up to 174 μN with room for higher values. The design, fabrication, and performance of the micro-emitters shown in this work can lead to real primary propulsion solutions for smallsats. The second part of this work utilizes the electrospray expertise to study the fiber initiation on electrosprays of high molecular weight polymer solutions, also known as electrospinning. First, I focused on the electrostatic jet initiation of a SU-8 polymeric solution studied with two different geometries in the so-called Near-Field regime to quantify the initiation parameters and illustrate the optimization of the electric field. Then, I functionalized suspended glassy carbon fibers derived from the same polymer solution using chemical vapor deposition. The temperature required for the deposition of WO3-x is generated by imposing a constant electrical current through the wire that causes joule heating. The deposition starts in the midpoint of the wire, extends to its ends as the current increases, and can be monitored in real-time by measuring the voltage drop across the wire. The resulting thickness and length of the coating are functions of the imposed current. This work showcases uniform and polycrystalline WO3-x coatings with thickness from 71 nm to 1.4 μm in glassy carbon wires with diameters between 780 nm and 2.95 μm. The same process is scaled-up by using carbon nanofiber mats, fabricated with far-field electrospinning, where the uniform temperature increase is homogenous to a good approximation by the dominance of radiative heat transport leading to very uniform WO3-x coatings. The functionalization of carbon micro nano materials described in this work can lead to novel, inexpensive, and bioavailable sensing solutions.

Electrospraying of nanodroplets modifies the target surface by sputtering its atoms and amorphizing the outer layers. Collisions of droplets of diameters between 1 nm and 1 m are not yet totally understood. Previous experimental studies have been performed and compared with Molecular Dynamics simulations using the pseudo-atom model (PA) for the droplets. This PA model considers each liquid molecule as a single particle with the same molecular mass. In this study a new atomic/molecule model (AM) which contemplates all atoms and bonds of the droplet is presented. Simulations of the collision of a 5 nm diameter droplet for formamide and EMIMIm are performed using both models. Droplet initial kinetic energies between 3 KeV and 15 KeV are tested. Results show that PA model collisions transfer almost all of their initial kinetic energy to the slab. Temperature fields after the collision are much higher for this model. Furthermore, highly energetic EMIMIm droplets pseudo-atoms penetrate further into the target whereas atoms in the AM model do not. The ratio of Silicon atoms sputtered and the amorphized region depth for the PA models are larger than for the atomic/molecule, which compare well to experimental results. The AM model is more accurate, especially when simulating droplets of high atomic mass for which the PA model particles are very energetic. Energy consumed to break bonds in the AM model is negligible in comparison with the droplet initial kinetic energy. Simulations show that unbreakable bonds can be used in the AM models obtaining almost identical results.

The electrospray ionization, also known as electro-hydrodynamic ionization, produces a beam of charged nanodroplets. These beams of nanodroplets can be used in material deposition, sputtering, and milling processes, all of which are currently done by beams of atomic and cluster ions. In electrospray atomization, a uniform beam of submicron-sized charged droplets is sprayed in vacuum in a strong electric field. The charged droplets are accelerated by the electric field induced by the electrodes. The space charge of the beam causes the droplets to repel each other and the beam to open up. In this dissertation, the development of an electrostatic column for electrospray beam focusing is presented. Coaxial electrodes were designed to extract and focus the beam. The beam is skimmed to agree with the paraxial assumption used in the projectile tracing. Fundamentals of the charged particle tracing under an electric field is discussed in detail and numerical simulations provide the projectile trajectory and image location where the beam is focused and energy density is maximized. These focused beams are used to sputter a semiconductor target. The performance of the focusing apparatus can be characterized based on the size and features of the impacted area. The error sources in lens system result in an increased image size appearing in the form of slightly sputtered area. These aberration figures are analyzed for different field strength to determine the size of imperfections in the apparatus and to be possibly corrected in next stages of development. With the exception of spherical and chromatic aberrations, other sources of errors can be eliminated or at least minimized if the misalignments between elements including electrospray source and electrodes are minimized. An innovative process is presented for fabricating the focusing lens using microfabrication techniques. This approach guaranties high precision and minimum misalignment between elements. Plasma activated bonding is used to permanently attach the silicon electrodes and glass standoffs. In addition to precision fabrication of apertures, the advantage of this approach over conventional machining is eliminating post fabrication assembly. Despite higher cost and difficulty of process, the microfabricated lens assembly successfully provides the voltage and accuracy requirements for a high resolution focusing lens.

The energetic bombardment of covalently bonded materials by electrosprayed nanodroplets causes sputtering, topographic changes and amorphization of the target's surface. The goal of this thesis is to investigate these phenomena using a variety of semiconductor materials and dielectric liquids. The electrosprays are characterized via time-of-flight spectrometry to determine the nanodroplet charge-to-mass ratio which, together with the acceleration voltage, yield the impact velocity, stagnation pressure, and kinetic energy of the projectiles. The damage caused by the beams on the surfaces of single-crystal targets such as Si, SiC, InAs, InP, Ge, GaAs, GaSb and GaN is characterized with different tools including a mechanical profilometer, an atomic force microscope, and scanning and transmission electron microscopy, measuring the sputtering yield (ejected atoms per projectile's molecule), sputtering rate, surface roughness, and the morphology of the surface affected by the beam. These figures of merit are quantified in terms of the projectile's size, molecular mass and kinetic energy. Nanodroplets are efficient sputtering projectiles, with maximum sputtering yields of 11.5 and 25.1 for the technological important but difficult to etch SiC and GaN. The maximum sputtering rates for SiC, GaN and GaSb are 720, 1750 and 2380nm/min. The surface roughness and sputtering yields typically increase with the projectile's kinetic energy for all targets, and exhibit sharp maxima for nanodroplets with high molecular mass. The very different sputtering by droplets that are macroscopically similar, and the strong dependence of the impact phenomenology on molecular mass, indicate that nanodroplet sputtering is intrinsically a molecular scale phenomenon, dominated by the transfer of energy under non equilibrium conditions, and hence not amenable to modeling with a continuum formulation. For the case of single-crystal silicon the influence of the projectile's velocity on amorphization is studied. The impacts of nanodroplets on a single-crystal silicon wafer at kinetic energies exceeding a threshold amorphatize a thin superficial layer, of thickness comparable to the droplet diameter. This thesis contributes to develop a new generation of energetic projectile sources in a previously unavailable particle size range, producing beams which can be electrostatically focused into submicrometric spots, similar to focused ion beams, or multiplexed for broad-beam batch fabrication.

Electrospray thrusters are unique among electric propulsion due to their inherent scalability and their potential to cover a broad range of specific impulses and thrust. Capillary emitters, in particular, offer advantages such as enhanced controllability and reliability over externally wetted and porous emitters. However, prototypes of capillary emitter electrospray thrusters have so far achieved modest specific impulses compared to other types of emitters. The maximum specific impulse of electrosprays is closely tied to the minimum flow rate at which cone-jets can emit stably. This study investigates how altering the geometry of capillary emitters can increase the specific impulse while maintaining the advantages that capillary emitters offer over externally wetted and porous emitters. Time-of-flight and direct mass flow rate measurements were conducted to characterize electrosprays of various ionic liquids in a vacuum. It was found that electrosprays with emitters with smaller diameters resulted in a lower stable minimum flow rate by more than an order of magnitude compared to those with larger diameters, leading to a two- to threefold increase in the maximum specific impulse.