24 Figure 1.4: Thrust versus delta-V tradeoff space for the intended use-case CubeSat fitted with the proposed propulsion system.. 76 Figure 3.1: Theoretical propellant density as a func
Small Satellite Technologies
The space era began with the launch of the first artificial satellites by the Soviet Union in 1957 and the United States in 1958, respectively These early satellites were small craft, marking the dawn of a new chapter in space exploration.
Sputnik 1 weighed 83.6 kg and Explorer 1 weighed 13.97 kg, with these early satellites’ modest masses reflecting the limits of 1950s launch vehicles rather than the ambitions of their creators In the following decades, heavy launch vehicles emerged, culminating in the Saturn V, capable of placing as much as 140,000 kg into low Earth orbit (LEO) This boost in launch capacity sparked rapid growth in satellite deployment for commercial, scientific, and national defense uses, driven by the space race The International Space Station, weighing about 420,000 kg, began construction in 1998, and its funding has been tentatively extended through 2030 Today, standard commercial communication satellites and flagship scientific missions like the James Webb Space Telescope weigh roughly 5,800 to 6,700 kg.
Over the past two decades, the trajectory of spacecraft size has shifted from continual growth to a resurgence of small satellites, enabled by the miniaturization of electronics and computing resources Today, NASA defines a small satellite as a spacecraft with a mass under 180 kg.
Small satellites offer several key advantages over larger ones, notably lower cost, easier access to space, and greater tolerance for risk There are three main costs in operating a satellite: design and construction, launch, and operation Because they are simpler and carry fewer systems due to size constraints, small satellites are typically cheaper to design and build, making them more cost-effective, though a 10 kg small satellite (CubeSat) developed by NASA can still cost in excess of $10 million In general, small satellites do not have the same service life as larger satellites, typically operating for months or years before retirement, often due to radiation effects.
Small satellites are more affected by orbital decay and atmospheric drag, which shortens their mission durations Their missions are therefore more succinct, in part because they lack the robust hardware and longer lifetimes of larger missions The most significant cost savings with small satellites come from their launch costs, which are substantially lower than those for larger spacecraft.
Large launch vehicles designed to place substantial payloads into orbit typically carry a mass margin—the extra mass beyond the primary payload that the vehicle can deliver This mass margin is often sold to small satellite operators as secondary payload opportunities The trade-off is that secondary payloads can only ride to the orbit of the primary payload or an intermediate orbit, but the arrangement yields significant cost savings on the launch Small satellites gain easier access to space, as government agencies such as NASA and the DoD regularly acquire mass margin on launches and allocate it among multiple customers Private organizations also leverage this approach to secure space access.
Small satellite programs tolerate higher risk due to their lower cost, enabling greater technical innovation and more flexible mission profiles The cost advantage relative to large flagship, state-sponsored missions reduces monetary risk and frees teams to experiment with new technologies and capabilities Consequently, small satellites routinely fly unproven hardware for technology demonstrations, whereas large missions typically rely on flight-proven components except during dedicated flight-test programs These dynamics have helped popularize the small satellite ecosystem over the past two to three decades, transforming the space industry by recognizing the value of smaller platforms.
A small satellite form factor has emerged that leverages the aforementioned values by introducing a standardized form factor, the CubeSat CubeSats enjoy standardization because their
CubeSat geometry is fixed to integer multiples of 10 cm cubes, with each 10 cm cube constituting 1U and a volume of 1 liter This standard unitization drives the development of universal satellite deployment mechanisms, or launchers, which in turn reduce launch costs by simplifying satellite integration with the launch vehicle and the deployment sequence Among the most popular CubeSat deployment systems is the Poly Picosatellite Orbital Deployer (P-POD), developed at California Polytechnic State University (Cal Poly).
[10], and the NanoRacks CubeSat Deployer (NRCSD) developed by a private launch services company, NanoRacks [11] Both devices, very simply put, contain CubeSats in a box during launch and then deploy them by opening a hatch on the box and pushing the satellites out via a large spring The simplicity of these deployment mechanisms, coupled with the relatively low cost of development, has led to an explosion in the adoption of the CubeSat form factor in academic, government, and commercial sectors Since CubeSats were first flown in 2003, approximately
1150 have been launched, according to an international organization who tracks their use [12] This popularity has led to the rapid growth of commercially available space hardware which is scaled down the development costs of small satellites, and especially CubeSats, and further propelled their popularity.
Propulsion System Fundamentals and Tradeoffs
CubeSat propulsion technologies have only recently matured enough to support meaningful adoption by designers NASA is actively tracking the development and maturation of a broad range of propulsion technologies—developed and marketed by numerous organizations—highlighting the diversity of options available A solid grasp of propulsion fundamentals provides the essential groundwork for exploring the breadth of propulsion systems currently in existence.
Propulsion systems have one specific job, to impart momentum on a spacecraft This is
5 always achieved via the principles of conservation of momentum and is stated in its most general form in Equation 1.2.1
Let Ṗ denote the time rate of change of momentum of a system It equals the sum of external forces acting on the system plus the sum of mass fluxes entering the system times their velocities, expressed as Ṗ = ∑F + ∑ṁv Some devices can alter momentum without mass flux by using solar radiation pressure from the Sun—these are solar sails, though they are not central to this discussion All other propulsion systems change momentum primarily by ejecting mass to produce thrust.
Delta-V, the amount of velocity change a propulsion system can impart, and thrust, which determines how quickly that delta-V can be achieved, are the most important figures of merit for propulsion These performance metrics cannot be maximized simultaneously due to inherent trade-offs in propulsion design A designer must optimize a propulsion system based on mission requirements and the relative cost of trading off delta-V against thrust.
Total delta-V achievable by a spacecraft depends on the propellant mass, the spacecraft’s total initial mass, and the propulsion system’s efficiency, expressed as specific impulse (Isp) This relationship is described by the Tsiolkovsky rocket equation, also known as the rocket equation In essence, delta-V is determined by the natural logarithm of the mass ratio (initial mass over final mass) multiplied by the effective exhaust velocity (Isp times g0); therefore, increasing propellant mass or Isp boosts delta-V for a given mass ratio, while a heavier spacecraft reduces it.
This equation is obtained by integrating the spacecraft’s acceleration, which depends on the propulsion system’s thrust and the evolving mass of the vehicle as propellant is burned It explicitly accounts for propellant consumption, capturing how the spacecraft becomes lighter during thrust and how this mass loss affects velocity The resulting relation is the rocket equation, shown here as Equation 1.2.2.
Here, Isp is the specific impulse of the propulsion system, a measure of thruster efficiency, g is Earth's gravitational acceleration, mi is the vehicle’s initial mass, and mf is its final mass after propellant is burned; the standard form of this relationship is Δv = Isp × g × ln(mi/mf), describing the maximum change in velocity achievable given the propellant mass fraction and engine performance.
When the propellant is fully expended, the vehicle’s final mass equals its dry mass, md, i.e., the mass of the vehicle with no propellant This final mass can also be expressed in terms of the dry mass md and the propellant mass mp, as shown in Equation 1.2.3, which relates the components of the vehicle’s mass before and after propellant burnout.
From this equation it can clearly be seen that to maximize the delta-V of a propulsion system, one should maximize specific impulse and propellant mass, and minimize vehicle mass
Specific impulse is a key metric that measures the efficiency of a propulsion system, representing thrust relative to the rate of propellant consumption This efficiency can be expressed in several forms, but Equation 1.2.4 provides the most meaningful expression for comparing propulsion performance By linking thrust, propellant mass flow, and effective exhaust velocity, Equation 1.2.4 serves as the standard reference in rocket propulsion analysis and design Understanding this relationship helps engineers benchmark different propulsion concepts and optimize system performance.
Equation 1.2.4 states that propulsion efficiency is directly proportional to the propellant's exit velocity The analysis assumes a perfectly collimated propellant mass flux, an idealization that is not realistic in practice For the purposes of this discussion, exhaust plume divergence and the velocity distribution of the propellant are not addressed A deeper analysis that accounts for these factors would typically produce a lower specific impulse than the ideal value defined by Equation 1.2.4.
Thrust is the second most important metric for a propulsion system In spacecraft design, the vehicle is assumed to operate in space, so the thrust-to-weight ratio is far less critical than for a launch vehicle that must lift off Earth Thrust determines how quickly the propulsion system can alter the total delta-V, making it a key factor in mission capability Although thrust can be expressed in several ways, the most relevant definition is the one shown in Equation 1.2.5.
𝑇 = 𝑚̇ 𝑣 (Equation 1.2.5) This states that thrust is equal to the mass flux of propellant times the exit velocity of the propellant Again, this assumes a columnated propellant flux which is a simplification
An essential factor in propulsion system design is the time required to reach the specified delta-V, since it directly influences orbital mechanics calculations In chemical propulsion, the delta-V is often treated as imparted instantly, a simplifying assumption that eases orbit-change calculations even though it is not physically exact By contrast, low-thrust propulsion can require days, weeks, or months to complete a delta-V maneuver, adding complexity to orbital calculations Therefore, propulsion time must be considered when modeling missions The thrust duration can be estimated by dividing the propellant mass by the propellant mass flux, as shown in Equation 1.2.6.
Within this dissertation, the thruster under consideration is an electrostatic thruster, described in more detail later The key thrust metric for this system is the effective accelerating voltage, defined as the line integral of the electric field along the path the propellant ion travels and distinct from the voltage applied to accelerating grids The relationship between the applied grid voltage and the effective voltage depends on factors such as thruster geometry, grid design, and plasma density; this transfer function can be simulated or measured, but it is outside the scope of this dissertation Therefore, when accelerating voltage is mentioned here, it refers to the effective voltage, not the grid voltage The accelerating voltage Va can be related to the kinetic energy of a singly ionized propellant via Equation 1.2.7, and for a specific propellant species this is described in Equation 1.2.8, where qe is the electron charge, Mp is the molar mass of the propellant species, and Na is Avogadro’s number The exit velocity can be solved for by equating the accelerated ion energy to the ion’s kinetic energy.
8 kinetic energy as in Equation 1.2.9 From this analysis we can then write exit velocity, specific impulse, and thrust in terms of system level variables in Equations 1.2.9 – 1.2.11 [1]
From Equations 1.2.10 and 1.2.11, maximizing accelerating voltage and minimizing propellant molar mass seem to be the most efficient way to achieve delta-V, but a tradeoff arises because the total thruster power must be considered The propellant current can be calculated from the propellant mass flux, the molar mass, Avogadro’s number, and the electron charge, as shown in Equation 1.2.12 Electric propulsion spacecraft are power-limited in how they operate the propulsion system, and because beam current is a deterministic value, Equation 1.2.12 can be solved for the propellant mass flux, as illustrated in Equation 1.2.13.
N_q (Equation 1.2.13) defines the effective propellant-beam power and shows that it can be calculated from the measured current and voltage The effective propellant beam power, as given by Equation 1.2.13, represents the portion of electrical input that contributes to the propellant beam It is important to note that this value is not the total power required to operate the thruster, since there are other elements in the system that dissipate energy This power calculation isolates the propellant-beam contribution from system losses, linking current, voltage, and beam power for propulsion analysis.
Power required by the propulsion system is the energy it must impart to the mass flux to achieve the desired thrust While the total system power is typically much higher than this minimum, this required-power measure provides a practical framework for mapping the thruster tradeoff space, linking thrust, efficiency, mass flow rate, and other propulsion-performance metrics.
𝑃 = 𝐼 𝑉 (Equation 1.2.14) Finally, the propellant mass flux can be solved for in terms of deterministic values and independent variables and is shown in Equation 1.2.15
Xenon Difluoride and Tungsten
Xenon difluoride (XeF2) is a solid crystalline compound that sublimes at room temperature, and, together with xenon tetrafluoride (XeF4), was first synthesized at Oak Ridge National Laboratory in the 1960s as one of the first known noble gas compounds In the following years, researchers studied the chemical and physical properties of XeF2 and XeF4 to refine nuclear materials for atomic weapons.
These compounds were proposed to react with heavy metal ions to form volatile fluorides, enabling easier separation from other ore materials for subsequent elemental and isotopic refinement The literature is unclear about how successful or useful this approach was, particularly given national security and strategic knowledge confinement concerns Nevertheless, it provides useful information on the nature of these compounds, including reported physical properties such as vapor pressure, density, and heat of sublimation, along with two distinct synthesis methods XeF2 did not find substantial commercial application until the semiconductor industry matured [19, 20].
There is little notable literature on commercial applications of XeF4, largely because vapor-phase XeF2 aggressively etches a range of materials with high selectivity over oxides, including silicon, tantalum, molybdenum, tungsten, and others, as reported in references [22–27].
Xenon difluoride (XeF2) is a highly effective etchant for materials that readily form volatile fluorides, enabling clean and selective material removal with minimal residues The strength of XeF2 lies in its distinctive etching mechanism, which drives the rapid formation and volatilization of fluorinated byproducts; this mechanism will be explored in depth in Chapter 3.
Low Temperature Co-Fired Ceramics
Low Temperature Co-fired Ceramics (LTCC) is a manufacturing technology akin to printed circuit board (PCB) fabrication, but it uses ceramic structural and dielectric layers instead of glass-epoxy laminates and deposits conductors with a silk-screened sinterable paste rather than etched copper sheets LTCC is commonly used to create packaging solutions for electronic components, offering superior performance under extreme conditions and enabling advanced features such as cavities and thick-films not found in PCBs These capabilities allow embedded circuit components—resistors, inductors, capacitors, and RF waveguides—to be integrated directly into the ceramic package.
LTCC technology is orders of magnitude more expensive than conventional PCB technology, so it has typically been reserved for extreme‑use or high‑reliability applications The ceramic material offers a very low dielectric loss tangent of 0.001–0.0014, which enables excellent performance in RF applications at higher frequencies Despite the higher cost, LTCC is chosen where low loss and reliable high‑frequency operation are critical.
At 10 GHz, dielectric losses become significant, but LTCC (low-temperature co-fired ceramic) offers chemical stability and non-reactivity that enable reliable operation in harsh environments with high humidity or chemical exposure LTCC ceramic devices boast a breakdown voltage well in excess of 4 × 10^10 V/m, making them well suited for high-voltage applications In addition, LTCC's structural and electrical properties remain stable up to temperatures above 500 °C, after which most electrical components would fail These properties position LTCC technology as an excellent choice for exotic applications where traditional printed circuit boards would not survive or meet performance requirements.
LTCC devices are built akin to PCB as previously stated Designs are created by stacking
Green tape comprises 13 individual layers of ceramic–polymer thick films, typically 0.005–0.020 inches thick and processed as 8" × 8" sheets Each layer can host vertical interconnects (vias) punched from the tape with a die and filled with conductive, sinterable paste, while lateral conductors are applied with the same paste via a silkscreen process Voids can also be punched or milled into the tape After processing, the layers are stacked and laminated under high pressure (2,000–4,000 psi) to form a multilayer stack, which is then co-fired at 850–1,000 °C to fuse the layers, burn off the polymer binders, and yield a single monolithic structure.
Current State of the Art in CubeSat / Small Satellite Propulsion
Propulsion systems can be categorized into three main sub-genres: chemical propulsion, thermodynamic propulsion, and electric propulsion Chemical propulsion relies on a chemical reaction or decomposition process to increase the enthalpy of a gas, which is then expelled through a nozzle, and it typically comes in two common variants: monopropellant systems that decompose in a catalyst-based chamber, and bipropellant systems that rely on a chemical reaction between two propellants in a combustion chamber Thermodynamic propulsion uses stored energy in the propellant that is released or transformed during use, with classic examples such as cold gas thrusters that discharge compressed gas through a nozzle, and devices like resistojet thrusters that add a heating element in the propellant expansion path Electric propulsion, also known as solar electric propulsion, uses electricity—usually generated by solar panels—to accelerate propellant, achieving propulsion by accelerating charged particles to high velocity and ejecting them through a nozzle.
14 by using electric or magnetic forces and the selection of which force is employed determines the sub-genre of electric propulsion
The current state of the art for propulsion suitable for CubeSats and small satellites is best summarized by NASA's 2018 report on the ecosystem of small satellite technologies This authoritative source highlights the propulsion options available for CubeSats, evaluates their performance, and situates them within the rapidly expanding small-satellite sector.
This periodically updated document serves as a benchmark for academia, industry, and government agencies, offering a clear snapshot of technology maturity and performance Table 1.1, drawn from this report and described in detail below, is organized by technology area and includes fields such as Product, thrust, specific impulse, and TRL status TRL, or Technology Readiness Level, is a nine-point scale used in aerospace to assess how mature a technology is and its readiness for deployment.
Table 1.1: Summary of the current state of the art for small satellite propulsion systems [9]
Propulsion System Types for Small Spacecraft
Product Thrust Specific Impulse (s) TRL Status
Cold Gas 10 mN - 10 N 40 - 70 GN2/Butane/R236fa 9
Vacuum Arc Thrusters 1 - 1300 àN 500 - 3000 Teflon 7, Titanium 7
Hall Effect Thrusters 10 - 50 àN 1000 - 2000 Xenon 7, Iodine 3
Ion engines with thrusts in the 1–10 mN range and propellants such as xenon or iodine are used in industry to describe the maturity of propulsion technologies [31] These assessments cover the spectrum from scientifically valid principles of operation to fully flight-ready and mature systems, with the TRL definitions shown in Table 1.2.
Table 1.2: Technology Readiness Level definitions
TRL Definitions TRL 1 Basic principles observed and reported
TRL 2 Technology concept and/or application formulated
TRL 3 Analytical and experimental critical function and/or characteristic proof-of-concept TRL4 Component/subsystem validation in laboratory environment
TRL5 System/subsystem/component validation in relevant environment
TRL 6 marks the demonstration of a system or subsystem model or prototype in a relevant end-to-end environment (ground or space) At TRL 7, a system prototype is demonstrated in an operational environment (ground or space) TRL 8 advances to the completion of the actual system, which is mission-qualified through testing and demonstration in an operational environment (ground or space) Finally, TRL 9 confirms the system as mission-proven through successful mission operations (ground or space).
Hydrazine is a monopropellant, meaning its decomposition products react with each other to produce thrust without needing to be mixed with an oxidizer Its chemical formula is N2H4, and its structure features two nitrogen atoms bonded by a single bond with two dangling hydrogens on each nitrogen, making it a highly toxic and unstable liquid at room temperature In a hydrazine thruster, the fuel flows over a catalyst bed where it decomposes to form hot gases, which are then expanded through a nozzle to generate propulsion This spontaneous decomposition under controlled conditions yields a propellant stream of high-velocity gas used in various spacecraft and propulsion systems.
Hydrazine thrusters are commonly used for attitude control on larger spacecraft, valued for their relative simplicity and the maturity of the technology, which supports a broad range of product offerings The main challenge with hydrazine systems is the difficulty of miniaturizing their support equipment, particularly redundant valving, which limits scalability and adds complexity for small spacecraft; in addition, high delta-V maneuvers are not practical due to the low storage density of hydrazine propellant These thrusters are typically characterized by strong thrust, good specific impulse, and a high TRL A CubeSat-oriented commercial example is Aerojet Rocketdyne’s CubeSat High-Impulse Adaptable Monopropellant Propulsion System (CHAMPS), a 1U system that carries up to 360 g of propellant and provides a thrust of 0.24 N.
– 2.9 N, dissipates ~ 2 W of power, and has a specific impulse of 215 s [32]
A new type of monopropellants, sometimes referred to as ‘green propellants’, have gained traction over the last decade They have been developed to address the toxicity concerns of hydrazine This makes it easier for a green propellant thruster to be integrated on small satellites or CubeSats Green propellants are like hydrazine in that they are a monopropellant which is an advantage for simplicity The challenge with green propellant systems is that they have low specific impulse, low storage density, and have miniaturization challenges These systems are not suitable for high delta-V maneuvers for the same reasons that other monopropellant technologies fall short, but they are an ever maturing technology The most mature example is the Busek BGT- X5 thruster which boasts 10% higher specific impulse and 45% greater storage density over hydrazine when using a proprietary green propellant (AF-M315E) This system takes up 1 U of volume and can provide 500 mN of thrust at a specific impulse of 220 – 225 s while dissipating
20 W of power during operation [33] Other noteworthy examples are the Busek BGT-X1 with 100 mN of thrust and a specific impulse of 214 s [34], the Aerojet Rocketdyne MPS-130 CHAMPS with 1.5 mN of thrust and specific impulse of 240 s [35], the Aerojet Rocketdyne GPIM Propulsion System with 400 – 1100 mN of thrust and a specific impulse of 235 s [36], and the ECAPS HPGP thruster with 1000 mN of thrust and a specific impulse of 232 s [37]
Cold gas thrusters are the simplest propulsion option, delivering thrust by expelling compressed gas through a nozzle This approach is a mature, widely used technology, especially on small satellites The main drawback is their inherently low specific impulse, resulting from the thermodynamics of an expanding gas, which limits cold gas thrusters to tasks like attitude control, reaction wheel desaturation, and minor station-keeping Consequently, they are ill-suited for significant delta-V maneuvers, though they remain reliable and easy to implement for small adjustments and stabilization needs.
Seventeen CubeSat thrusters are generally characterized by good thrust, low specific impulse, and a high TRL One of the most technologically mature cold gas thrusters for CubeSats is the VACCO Micro CubeSat Propulsion System, which flew on two interplanetary missions launched in May.
MarCO A and MarCO B, the 2018 Mars‑bound CubeSats, performed Mars flybys in November 2018 Their propulsion system delivered a total impulse of 755 N-s with a thrust of 25 mN and a specific impulse of 25 s Other notable micropropulsion examples cited include the SSTL SNAP 1 thruster with 50 mN of thrust and a specific impulse of 43 s, and the UTIAS‑SSFL CNAPS thruster, which is also referenced in the discussion.
40 mN of thrust and a specific impulse of 35 s [40], Microspace Rapid POPSAT-HIPI thruster with
1 mN of thrust and a specific impulse of 43 s [41], the GOMSpace MEMS Cold Gas thruster with
1 mN of thrust and a specific impulse of 50 – 75 s [42], and the VACCO Industries CPOD with 25 mN of thrust and a specific impulse of 40 s [43]
Resistojet thrusters are a variant of cold gas propulsion that place a resistive heater in the propellant stream to add thermal energy, raising the gas enthalpy and exit velocity, and thus increasing specific impulse This efficiency gain comes at the cost of higher spacecraft power demands A notable flight example is NovaSAR (2012), which used a xenon-based resistojet for attitude control The LPR thruster from SSTL delivered 18 mN of thrust with a specific impulse of 48 s while dissipating 30 W Other noteworthy resistojet examples include the CU Aerospace PUC thruster (5.4 mN, Isp 65 s), the CU Aerospace CHIPS thruster (30 mN, Isp 82 s), the Busek AMR thruster (10 mN, Isp 150 s), and the USC FMMR thruster, which provides 0.13 mN of thrust.
Pulsed plasma thrusters (PPTs) and vacuum arc thrusters offer very high specific impulse by using a high-voltage arc to ablate a small amount of solid propellant, ejecting it at high velocity Because the propellant is stored and consumed as a solid, delivering more propellant to the arc region—especially in low gravity—is challenging A common design pushes a cylindrical slug of solid propellant (such as Teflon) through a tube to the discharge section with a spring, but the spring’s travel limits the feed As a result, this technology is well suited as a reaction control thruster or to desaturate reaction wheels, but not for high delta-V maneuvers due to the extremely limited propellant mass The most mature commercial option is the Busek BmP-220 Micro-Pulsed Plasma Thruster, a compact 0.5 kg device delivering 175 N·s of total impulse and occupying a system volume of 0.375 L.
Micro-thrust propulsion features compact thrusters that deliver very small thrust with high specific impulse W provides 0.14 mN of thrust with a specific impulse of 536 s [49] Notable examples include the Busek MPACS thruster delivering 0.14 mN of thrust and a specific impulse of 830 s [50], the Primex Aerospace EO-1 PPT thruster offering 0.14 mN of thrust and a specific impulse of 1150 s [51], and the George Washington University àCAT thruster with 0.02 mN of thrust and a specific impulse of 3000 s.
[52], Würzburg University UWE4 Arc Thruster with 0.01 mN of thrust and a specific impulse of
Proposed Propulsion System Architecture
This dissertation proposes a propulsion system architecture and assesses its feasibility through fundamental research designed to generate engineering data for building a prototype, or at least to identify the key areas that require further investigation before prototyping The architecture is motivated by a defined use-case to address a hypothetical need, and its conceptual layout is illustrated in Figure 1.3.
The proposed propulsion system is a heavy metal subliming electrostatic propulsion system, a novel architecture not previously reported in the literature The novelty arises from its unique propellant generation paradigm and a new type of thruster By leveraging controlled sublimation of a heavy metal to generate propellant and electrostatic acceleration, the concept offers distinctive performance and propellant handling characteristics This combination sets the design apart from conventional electrostatic propulsion approaches and represents a fresh contribution to the field.
Figure 1.3: Schematic of the proposed propulsion system
This article describes a propellant generator composed of two interconnected chambers: a subliming chamber storing a corrosive material and an etching chamber housing a heavy metal The corrosive material sublimates and etches the metal to form a dense propellant, while a thermal control system regulates the temperatures of both chambers and a valve separates them to control flow By storing propellant in a maximally dense form, this arrangement, as has been shown theoretically, helps optimize the achievable delta-V of the propulsion system.
This dissertation analyzes the corrosive materials XeF2 and tungsten (W) as reactants that produce a stream of Xe and WF6 gases, which can be used as a propellant The same reaction scheme also applies to XeF4 and W, with the theoretical potential for slightly better performance This approach offers the highest propellant storage density reported to date, with a theoretical maximum of 5.44 g/cm^3 for XeF2 and W and 5.70 g/cm^3 for XeF4 and W XeF4 was not explored in this research because it is not readily available commercially.
The propellant generator is separated from the electrostatic thruster by a valve, while the electrostatic RF ion thruster comprises a gas flow regulator, power and control electronics, and the thruster body The gas flow regulator sets the propellant mass flow rate into the thruster body, a critical operational parameter for propulsion systems The electronics coordinate valves, flow control, and DC and RF power delivery to the thruster, enabling precise control of ion production and thrust The thruster body ionizes propellant and accelerates it to produce thrust, following a classical electrostatic RF ion thruster design The system is built with a novel LTCC (low-temperature co-fired ceramic) process and materials system, which embeds electrodes in a tough, chemically resistant ceramic to enhance grid lifetime The LTCC manufacturing technique also allows for very efficient packing of components.
To enhance compact production, this work examines 26 functional elements of a ceramic thruster design, highlighting how ceramic materials enable very high voltages and temperatures—limited only by the drive electronics and ancillary equipment The research focuses on the behavior of three novel propulsion-system components—the sublimation chamber, the etching chamber, and the thruster body—while noting that heaters, valves, flow-control hardware, and electronics are established engineering elements and not the primary academic focus.
To avoid arbitrary performance targets, the proposed propulsion system was assessed against a concrete use-case The hypothetical mission was an interplanetary 3U CubeSat requiring a delta-V exceeding 1000 m/s The design constraints included a 10 W limit on propellant-beam power and a propellant storage volume of only 0.1 L This profile, chosen in 2014, was based on the assertion that achieving such performance would be exceedingly difficult with the technology available at the time; the sole competing option was an iodine-propellant propulsion system under development then That mission has been delayed due to propulsion-system development challenges caused by propellant corrosion It was originally intended to provide a 200 m/s delta-V maneuver to lower its orbit from a 600 km circular orbit to 300 km.
Exploring the trade-off space defined by Equations 1.2.16 and 1.2.17 for the target use-case, the method selects the intended power, propellant, and propellant storage volume In this analysis of a power-limited, volume-constrained spacecraft such as a CubeSat, the chosen propellant determines both the average propellant storage density and the average molar mass, which in turn set the thrust and specific impulse as functions of the effective accelerating voltage; a plot showing thrust and Isp versus voltage is then produced to visualize the relationship.
As shown in Figure 1.4, the calculations indicate that an applied accelerating voltage of 318 V would yield a delta-V of 1000 m/s and a net thrust of 106 µN This delta-V would require approximately 112 days to achieve on the CubeSat use-case, establishing the performance target for this work.
Figure 1.4: Thrust versus delta-V tradeoff space for the intended use-case CubeSat fitted with the proposed propulsion system
This study investigates the sublimation of XeF2 as a function of surface area and temperature, with results presented in Chapter 2 It also analyzes the etching behavior of XeF2 on tungsten (W), as described in Chapter 3 The design, fabrication, and testing of the LTCC electrostatic thruster are detailed in Chapter 4 The conclusions and a discussion of future work and potential research directions arising from this dissertation are presented in Chapter 5.
Accelerating Voltage (V) Specific Impulse versus Thrust - Tradeoff Space
Sublimation Dynamics of Xenon Difluoride
Methodology and Experimental Apparatus to Study Sublimation Dynamics of XeF 2
An experimental apparatus and procedure for investigating sublimation dynamics were proposed by the author and approved by the committee Figure 2.1 shows a schematic of the physical test setup The system comprises a vacuum chamber, a crystal sample holder, a pressure transducer, a pneumatic vacuum vent valve, and an in-situ thermocouple probe to monitor the chamber interior, with the vacuum chamber itself designed as a custom component to support precise measurements.
Figure 2.1 shows the sublimation dynamics experimental setup, with hardware that was fabricated in-house The vacuum chamber body is a large block of 6061-T6 aluminum, providing mounting points for a vacuum line, valves, and pressure transducers, as well as a substantial thermal mass to enhance thermal stability during operation A pocket milled into the block forms the interior cavity of the chamber, while a lid—also machined from 6061-T6 aluminum—seals to the chamber body with eight 1/4"-28 socket-head cap screws and double o-ring face seals to ensure a reliable seal.
The abundance of high-thread-pitch bolts lets the lid be clamped very tightly without risking galling the threads over many opening and closing cycles, while ensuring a reliable o-ring seal All other vacuum connections were 1/8 inch.
Using NPT or 1/4” compression, Swagelok-style fittings provide excellent vacuum sealing thanks to their tapered design, while eight bolts apply about 4,000 pounds of clamping pressure on dual O-rings at the chamber lid to create a robust vacuum seal with leak rates in the torr/second range The crystal holder is an aluminum block measuring 0.5" x 0.9" x 0.9" with four through-holes of nominal diameters 0.050", 0.075", 0.100", and 0.125", chosen to yield varying XeF2 crystal surface area; during experiments, only one hole was filled at a time to vary surface area in the experimental design The vacuum chamber's internal volume, including dead space in the transducers, was estimated at 44.8 cm^3 based on precise CAD models.
Pressure measurements were provided by one of two Baratron pressure transducers to achieve a wide dynamic range: the first transducer has a maximum range of 10 torr with an accuracy of ±0.001 torr, and the second has a maximum range of 100 torr with an accuracy of ±0.010 torr Each device uses a calibrated Inconel diaphragm fitted with a factory-calibrated strain gauge, with pressure determined from the manufacturer’s proprietary calibration curves based on diaphragm deflection, and readout is supplied by an MKS 651B transducer controller The pneumatic vent valve is normally off and opens when supplied with 1–3 bar of air pressure; supply air is provided by a solenoid valve plumbed to house air, driven by a high-power op-amp voltage follower and an analog output from the data acquisition (DAQ) system A type-K thermocouple enables real-time in-situ temperature measurements and is read by an AD595 hermetic thermocouple driver with internal temperature compensation, the device’s output being a voltage proportional to temperature.
31 with a 10 mV/°C slope Additional hardware involved in the sublimation dynamics setup was a vacuum pump and an environmental control chamber
An environmental control chamber was used to maintain precise experiment temperatures The thermocouple driver and the solenoid op-amp were powered from ±15 V supplied by LV7815 and LV7819 regulators, themselves fed by a benchtop power supply A custom LabVIEW program developed for this study orchestrated the experiment, controlling the valve and sampling thermocouple temperature via the NI USB-6216 USB data acquisition device The pressure transducer was powered and read by the MKS 651B controller, which communicates with the LabVIEW host computer over a serial interface.
A photograph of the sublimation dynamics experimental setup is shown in Figure 2.2
Figure 2.2: Photograph of the sublimation dynamics experimental setup inside the environmental control chamber.
Pilot Data and Initial Observations
An upfront validation of the experimental test apparatus, procedure, electronics, and software was conducted before the main test campaign to identify and address potential concerns The first three sublimation trials (Trials 1–3) failed due to data corruption, prompting fixes in the LabVIEW code to ensure proper data storage Trials 4–7 then produced clean data with no instrumentation or software errors, and these results were serially analyzed Based on the analysis, modifications to the test setup and experimental method were implemented, yielding a vetted and reliable experimental method and test setup for successful Trials 4–7.
Trial 4 used an experimental method that began by loading approximately 75 mg of XeF2 into a 0.075‑inch hole in the crystal holder, tamping it down with a precisely ground 0.075‑inch drill rod made of hardened tool steel, and sealing the bottom with Kapton tape to prevent crystals from dropping out during handling The crystal holder was then placed in the vacuum chamber, the apparatus sealed, and the system pumped down The process was software-controlled: the vacuum vent valve was opened until the start-pressure threshold of 0.3 torr was reached, at which point the valve was closed and the pressure and temperature were recorded as a function of time for 10 minutes at a defined sampling frequency.
Operating at about 10 Hz, the experiment collected data for a 10-minute period and saved the samples to an Excel file, marking one sublimation cycle; after completing the first cycle, the process was repeated by opening the vacuum vent valve until the initial experiment pressure was reached, then closing the vent and recording pressure and temperature data again, with this sequence carried out for 50 cycles.
Thirty-three cycles were performed, a number chosen not strictly out of necessity but to guarantee complete consumption of XeF2 in the trial This approach formed the foundation for all later sublimation experiments, and the final procedure retained the core elements of the original protocol with only minor modifications.
The results from Trial 4 were simultaneously interesting and concerning Thirty-five of the
In a XeF2 sublimation study, 50 cycles had sufficient XeF2 mass to sublime until the calculated vapor pressure was reached or exceeded, and the raw time-based pressure traces shown are only for trial 4 as representative of the overall data The first issue was that every cycle reached a maximum pressure higher than the calculated vapor pressure; Figure 2.3 shows the maximum pressure reached and the vapor pressure computed from the average temperature [16] The second issue was that the sublimation time constant, τs, was not constant but increased with time; τs was defined as the time to reach 95% of the calculated vapor pressure divided by three, an approach that assumes first-order kinetics (dP/dt proportional to P); this time constant is shown in Figure 2.4 The third issue was that the pressure traces did not exhibit the expected asymptotic behavior, since theory predicts the pressure should stabilize at the vapor pressure, producing a horizontal asymptote The fourth issue was that the maximum pressure reached was not constant and decreased with time, being larger than the vapor pressure but not constant Additionally, Cycle 1 displayed curious behavior and did not follow the smooth curve observed in the other traces Pressure traces from cycles 1–5 are shown in Figure 2.5 and traces from cycles 1–35 are shown in Figure 2.6.
The chief goal in analyzing the data from Trial 4 was to determine if the observed effects
34 described above were real and reflecting the true sublimation dynamics of XeF2 or if there were issues with the method or apparatus which led to unexpected results The analysis below led to introducing three modifications to the experimental method and are described in detail in the next several paragraphs
The first modification was to address pressure overshoot, the behavior of pressure rising significantly higher than the calculated vapor pressure (Figure 2.3) This effect was hypothesized
Figure 2.3: Maximum pressure compared to the calculated vapor pressure of XeF2 for Trial 4
Trial 4: Maximum Pressure vs Cycle Number
Trial 4 DataCalculated Vapor Pressure
Figure 2.4: Time constant of sublimation for Trial 4 over 35 cycles
Figure 2.5: Pressure traces from sublimation cycles 1 – 5 for Trial 4
Cycle Number Trial 4: Time Constant vs Cycle Number
Cycle 1Cycle 2Cycle 3Cycle 4Cycle 5
Figure 2.6 presents pressure traces from sublimation cycles 1–35 for Trial 4, where outgassing of absorbed water inside the vacuum chamber was implicated as the cause of the observed anomalies The atypical behavior seen in Trial 4, Cycle 1 (Figure 2.5) was interpreted as a similar transient To address these issues, a bake-out step was introduced before Trial 5: heating the environmental control chamber to 60 °C for 8 hours to drive off absorbed moisture The system could tolerate higher temperatures, but a cautious approach was taken to protect the pressure transducers from damage The bake-out eliminated the Cycle 1 odd behavior but showed no noticeable effect on the pressure overshoot.
To address the increasing time constant of sublimation observed in Figure 2.4, the second modification targeted the factors influencing sublimation kinetics The effect was hypothesized to arise from changes in the microscale surface area of the XeF2 crystals, which can influence surface reactivity and mass transfer at small scales During sample preparation, the XeF2 crystals were loaded into the sample holder, a step that involved tamping down the crystals to ensure stable packing and reproducible contact with the holder.
Time (min)Trial 4: All Pressure Traces
Testing with 37 crystals on a drill rod revealed mechanical weakness, leading to fragmentation where some crystals pulverized to a very fine size while others persisted as larger pieces The results are qualitative, as there was no method to measure the actual in-situ particle-size distribution.
An apt analogy for this process is to imagine taking all the dishes in your home and dumping them into a single large box If you lay a large flat board across the top and compress the stack uniformly, some dishes will break while others remain intact Although the crystals may start with a particular size distribution, the compression will produce a final distribution containing fragments significantly smaller than the original crystals and some particles that stay at the original size This illustrates how crystal packing occurs in the sample holder and is important for understanding the resulting sublimation dynamics.
Sublimation is the phase transition from solid to gas that occurs when a material is below its triple point, with deposition acting as the complementary process The sublimation rate and dynamics depend on surface area because the solid–gas interface governs the equilibrium and exchange between phases In the experiment, surface area is varied as an independent variable by using crystal holder holes of different sizes, demonstrating how macro- and micro-scale surface area influences sublimation dynamics The observed time constant for sublimation was shorter in earlier cycles and longer in later cycles, a pattern explained by changes in the crystal size distribution: earlier cycles included more small crystals, while later cycles favored larger crystals Since the smallest crystals have the greatest surface area-to-volume ratio, they sublimate first and more rapidly, driving the initial fast sublimation phase.
A 38-to-volume ratio was used, which meant that the crystal size distribution was not constant over the course of the experiment and represented an uncontrolled and unmeasurable variable that affected the experimental results This variability was evident in every single trial conducted throughout the study, highlighting the pervasive influence of crystal growth dynamics on the outcomes.
An essential realization emerged that helps interpret the results of the sublimation dynamics experiment and stands as its central takeaway and conclusion: the size distribution of crystals is a key factor governing the sublimation dynamics of XeF2 This finding clarifies why crystal-size variability influences sublimation behavior and highlights the need to account for size distribution in future studies.
Consideration of the depth of the holes in the sample holder as a factor contributing to the variance in the sublimation time constant suggested that as XeF2 crystals sublimate, their surface would move deeper into the hole; for this to influence the time constant, gas flow from the crystal surface to the vacuum chamber would have to be choked by the hole, but this possibility was ruled out because the geometric change would be negligible compared with the much more tortuous paths for expansion and diffusion of gas throughout the experimental setup; therefore, the observed variation in the sublimation time constant is attributed to the size distribution of the XeF2 crystals and a random variable within the study, representing a reinterpretation of the data rather than any modification to the experimental process or setup.
The third modification targeted the absence of asymptotic behavior in the pressure versus time traces (Figure 2.5 and Figure 2.6) Initially, this anomaly was attributed to a combination of outgassing from absorbed water and an air leak from the external environment into the experimental setup Rather than approaching a steady asymptotic value, the pressure exhibited a linear increase with time after transient changes in sublimation during the first ~4 minutes of the experiment.
Sublimation Dynamics Full Factorial Study
Pilot studies validated the experimental method and apparatus used to study the sublimation dynamics of XeF2 A full factorial study was conducted to quantify the influence of temperature and macroscale surface area, represented by crystal holder size, on sublimation dynamics The temperature factor ranged from 20–50 °C in 10 °C steps, while the crystal holder diameter ranged from 0.050–0.125 inches in 0.025-inch steps All 16 factor combinations were studied by loading the crystal holder with XeF2 and conducting sublimation dynamics measurements in accordance with the pilot-study guidelines Table 2.2 shows the nominal values of the independent variables.
Table 2.2: Table of nominal independent variable values for the 16 trials in the sublimation dynamics experiment
Trial Number Temperature (̊C) Sample Holder Diameter
This study treats the effluence of XeF2, defined as the mass flow rate during sublimation, as the primary dependent variable Each experimental trial received the same analytical treatment as the pilot trials The maximum pressure observed was recorded and the sublimation time constant was computed Additional analyses calculated vapor pressure and effluence from the pressure traces for each cycle and each trial Vapor pressure was estimated in two distinct ways, and the effluence rate as a function of chamber pressure was determined using four different methods The maximum pressure, the sublimation time constant, the vapor pressure, and the estimated effluence were then plotted against the independent variables of temperature and sample holder diameter.
The most impactful result of the sublimation dynamics study is the estimate of XeF2 effluence as a function of chamber pressure Two separate models were developed to predict effluence as a function of temperature, and effluence was found not to be significantly dependent on sample holder diameter This insight is especially relevant to the intended XeF2 sublimation use case, where the propellant mass flowrate directly affects thrust in a propulsion system Because XeF2 effluence constitutes part of the total propellant mass flowrate, it must be characterized over a reasonable range of operating parameters Overall, effluence was found to depend on both chamber pressure and temperature.
During the sublimation dynamics experiment, the environmental control chamber is first set to the trial's nominal temperature and the system is allowed to equilibrate for several hours At each temperature, the pressure transducer is calibrated by pumping the chamber down to a base pressure well below the transducer's sensitivity threshold, followed by nulling the pressure controller to ensure accurate readings This calibration procedure ensures reliable measurement of sublimation dynamics across the temperature range.
Samples were loaded with XeF2 crystals in the appropriate sample hole and the crystal mass recorded after following the transducer manufacturer’s recommendations; the sample holder was then placed in the test chamber, allowed to reach thermal equilibrium, and the vacuum was applied with the run configured in the software The experiment used 50 cycles, each with a 15-minute cycle time and a 5-minute thermal stabilization wait between cycles, more cycles than strictly needed to sublimate all XeF2 crystals to ensure efficient material use During each cycle, pressure, temperature, and time were recorded at 20 Hz and stored in separate Excel files for subsequent processing A custom Matlab code analyzed the data to extract the maximum pressure, the sublimation time constant, and calculated vapor pressure and effluence; the code also produced a down-sampled pressure-versus-time dataset for each cycle because each cycle contained ~15,000 data points, which would be cumbersome to graph, with the down-sampled traces containing ~300 points that yielded smooth plots without aliasing.
The time constant for sublimation was first estimated by identifying the time when the pressure reached 95% of the calculated vapor pressure and dividing that time by three This method assumes a first-order sublimation process, an approximation that, as discussed earlier, is not entirely correct, but it nonetheless provides a reasonable estimate of the time constant The sublimation time-constant results are presented by temperature, summarized in four plots; each plot includes results from every cycle and each sample holder diameter corresponding to that temperature.
Across successive cycles of the experiment, the sublimation time constant increases with the cycle number, reflecting evolving sublimation dynamics as the trial sequence progresses Figures 2.13 through 2.16 illustrate the temperature-dependent results at 20 °C, 30 °C, 40 °C, and 50 °C, respectively, while Figure 2.17 presents the average time constant for each trial The complete time-constant data are provided in Table 2.3.
The time constant for sublimation was a situation specific result The time constant first
Figure 2.13: Time constant of sublimation for trials conducted at 20 ̊C for four different sample holder diameters
Figure 2.14: Time constant of sublimation for trials conducted at 30 ̊C for four different sample holder diameters
Figure 2.15: Time constant of sublimation for trials conducted at 40 ̊C for four different sample holder diameters
Figure 2.16: Time constant of sublimation for trials conducted at 50 ̊C for four different sample holder diameters
Figure 2.17: Average time constant of sublimation as a function of sample holder diameter at four different temperatures
Effluence Diameter (mil) Average Time Constant vs Effluence Diameter
Table 2.3: Summary of all time constant of sublimation data
It was previously assumed that sublimation dynamics followed a first-order process, but this assumption is incorrect The associated time constant is only relevant to the experimental test chamber, whose CAD-estimated volume is 44.8 cm^3 This timing provides a sense of how long a XeF2 propellant storage vessel would take to pressurize after venting, but in spacecraft operations the time constant is typically negligible because a thruster using this propellant would operate for hours, days, or weeks at a time.
The propulsion system for the intended use case is inherently low-thrust, which means achieving a meaningful change in velocity would require a substantial amount of time Consequently, the development horizon for any noticeable delta-v is extended Additionally, the sublimation time constant should scale linearly with the volume of the propellant vessel, implying that larger tanks would exhibit longer sublimation times and influence overall propulsion performance.
Sublimation in a vented propellant vessel is temperature dependent, with a generic time constant for sublimation of the vessel volume estimated at 2000–3600 s/L across the measured temperature range This value is obtained by dividing the average sublimation time constant by the chamber volume The time required for a vented propellant vessel to return to the calculated XeF2 vapor pressure is about three times the time constant, yielding a pressurization time of roughly 1.7–3.1 hr/L The practical impact of this rate depends on whether the sublimed gas is consumed in a continuous or pulsed configuration and on the required propellant mass flow rate.
In the sublimation dynamics study, the maximum pressure observed in each trial differed from the calculated vapor pressure of XeF2 [16] Each trial, defined by a cycle number and a given sample holder diameter, produced a distinct maximum pressure Plots of maximum pressure versus cycle number show that the pressure reached in a trial tends to decrease as cycle number increases These maximum-pressure plots are presented for each temperature investigated and for each sample holder diameter, with the theoretical vapor pressure determined by the temperature included as a reference The maximum pressure attained for each cycle is shown in Figures 2.18 through 2.21 at 20 °C, 30 °C, 40 °C and 50 °C, respectively; each figure contains results from all four sample holder diameters.
An estimate of vapor pressure at each temperature, independent of cycle number or sample holder, was obtained using two complementary approaches In the first approach, the maximum pressure reached in each cycle was averaged, considering only cycles with sufficient XeF2 remaining to produce a measurable rise In the second approach, the peak pressures were averaged again, but the data were truncated by removing the first two cycles, which were treated as outliers.
Cycle 59 exhibited significantly higher peak pressure than the other cycles due to transient effects These transients arose from a combination of absorbed water released from air out-gassing and an elevated sublimation rate caused by the widest distribution of crystal sizes present during those two trials The relative contribution of these two transient effects was unknown, but its exact contribution remained unresolved.
Figure 2.18: Maximum pressure reached at trials conducted at 20 ̊C for four different sample holder diameters; theoretical vapor pressure of XeF2 at 20 ̊C
50 mil 75 mil 100 mil 125 mil Theoretical Pv
Figure 2.19: Maximum pressure reached at trials conducted at 30 ̊C for four different sample holder diameters; theoretical vapor pressure of XeF2 at 30 ̊C
Figure 2.20: Maximum pressure reached at trials conducted at 40 ̊C for four different sample holder
50 mil 75 mil 100 mil 125 mil Theoretical Pv
50 mil 75 mil 100 mil 125 mil Theoretical Pv
61 diameters; theoretical vapor pressure of XeF2 at 40 ̊C
Figure 2.21 shows the maximum pressure reached in trials conducted at 50 °C for four different sample holder diameters, alongside the theoretical vapor pressure of XeF2 at 50 °C It was hypothesized that crystal size distribution was the stronger effect than outgassing, and this outlier behavior became very clear when examining the plots of maximum pressure versus cycle number The first estimate of vapor pressure was referred to as the ‘average vapor pressure’ and the second estimate was referred to as the ‘truncated average vapor pressure,’ in addition to a calculated vapor pressure The average vapor pressure, truncated average vapor pressure, and calculated vapor pressure are plotted versus temperature in Figure 2.22, and the results of the vapor pressure estimates are summarized in Table 2.4.
The vapor pressure measurements in this study are important for two reasons: first, they enable direct comparison with existing literature, and second, they provide an estimate of the maximum operating pressure a propellant vessel may experience in the intended application This maximum propellant pressure is a critical factor in assessing system performance and safety in the reported use case.
50 mil 75 mil 100 mil 125 mil Theoretical Pv
62 in the design and operation of a propulsion system because it is head pressure that causes propellant
Figure 2.22: Plot of estimated and calculated vapor pressures as a function of temperature
Table 2.4: Summary of the vapor pressure results for all trials conducted
Truncated Average Vapor Pressure (torr)
XeF 2 Vapor Pressure vs Temperature
Average Vapor PressureTruncated Average Vapor PressureTheoretical Vapor Pressure
Sublimation Dynamics Conclusions
Conclusions from the sublimation dynamics study show that XeF2 effluence can be predicted by a model when operating conditions lie within the linear range defined by the upper and lower bounds in Figure 2.35, and the study identifies several factors that influence the effluence rate, with the crystal size distribution of XeF2 being the most important Microscale crystal surface area has a significant impact on effluence, while macroscale surface area is not significant because microscale effects dominate Another key finding is that sublimation does not stop at XeF2 vapor pressure, due to the molecule dissociating into Xe and F2 gases.
This study aims to elucidate the sublimation dynamics of xenon difluoride (XeF2) to inform the design of a propellant delivery device XeF2 is widely used as a subliming etchant in the electronics industry, yet its behavior for the proposed propulsion-related application has not been studied in depth While present applications rely on XeF2’s established etching properties, this work concentrates on the parameters that will govern a reliable XeF2-based propellant supply By examining sublimation rate, stability, and supply under conditions relevant to the intended use, the study highlights the key knowledge gaps and lays the groundwork for engineering a viable delivery system.
Bounds for the Linearized Effluence Model
XeF2 sublimation dynamics were explored within a timed dosing framework tuned to yield reliable etching results, an approach that remained entirely empirical and dependent on the characteristics of a single etching tool or process The data presented in this dissertation represent the only empirical examination of the nuances of XeF2 sublimation to date, and the findings are expected to be valuable for future designers of use-case propellant delivery systems for small satellites.