University of Southern California Dissertations and Theses

Experimental investigation on dusty surface charging in plasma

(USC Thesis Other)

Experimental investigation on dusty surface charging in plasma

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content EXPERIMENTALINVESTIGATIONONDUSTYSURFACECHARGINGIN PLASMA by Kevin L. Chou A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY (ASTRONAUTICAL ENGINEERING) December 2017 Copyright 2017 Kevin L. Chou ii Dedication ...To my family and all who have supported me iii Acknowledgements Though the path to this dissertation has come with many bumps along the way, the past six years have been extremely rewarding and have significantly increased my ability to apply basic concepts to solve complex problems. It is the hardships, both academically and per- sonally, and solutions to those problems that I will take with me as I begin my career. I owe thanks to numerous individuals who have helped me along the way. First and foremost, to my advisor, Dr. Joseph Wang, thank you for constantly guiding and supporting me. You have provided me with a strong foundation in plasma physics, and the conversations we have had over the years regarding my experimental results have significantly increased my critical thinking abilities. If it were not for you, I would not be the scientist I am today. I am forever grateful to have had the opportunity to work at the University of Southern California’s Laboratory of Astronautical Plasma Dynamics To the members of my qualifying exam and dissertation committee, Dr. Mike Gruntman, Dr. Edward Rhodes, Dr. Dan Erwin, and Dr. Joseph Kunc, thank you for your time and continual support. It is with your time and outside perspective that helped guide my research. To Dr. Mengu Cho, thank you for opening the doors to your laboratory and teaching me the fundamentals of experimental plasma physics. Without the knowledge I gained at KIT, I would not have been able to obtain the experimental results presented in this dissertation. To Dr. Henry Garrett, thank you for taking the time to speak to me about dusty plasmas and providing me with space suit material samples. Your assistance allowed me to close the loop on my research. iv To the previous and current members of the Laboratory of Astronautical Plasma Dynam- ics, thank you all for being so influential throughout my journey. To Dr. John Polansky, thank you for teaching me the ropes in the lab. To Will Yu, thank you for being such a great lab partner these past six years and working through all the problems we’ve come across. To Dr. Daoru Han, thank you for assisting me in my numerical simulations and always taking the time to help me out. Thanks to Dr. Scott Hughes, Yuan Hu, Yinjian Zhao, and Daniel Depew for the helpful discussions on plasma physics. To the ASTE staff, thank you Dell, Ana, Marrietta, Tina, Norma, Linda, and Hayley for all the help you have provided me while I was in the department. Finally, I want to thank my parents and family for their continuous support and sacrifice. They constantly motivated me to work harder and dream bigger. To Denise Chou, thank you, thank you, thank you. This research was supported in part by NASA LASER program grant NNX11AH21G and a fellowhip by USC Viterbi School of Engineering. v Abstract The objective of this dissertation is to study the effects of dust coverage on in plasma through experimental, analytical, and numerical investigations. Specifically, this dissertation investigates: 1. the effects of dust coverage on plasma charging of both conducting and insulating surfaces under plasma conditions similar to that at the lunar terminator 2. theeffectsofdustcoverageonphotoelectronchargingofbothconductingandinsulating surfaces, similar to that on the lunar dayside 3. the effects of dust coverage on plasma charging under multibody-plasma interaction conditions To study plasma charging at the lunar terminator, an electrostatic gridded ion thruster was used to generate a mesothermal plasma, which impinged the target surface at a zero degree angle of attack. Due to the lack of viable methods to directly measure dielectric surface potentials in plasma, a novel dielectric surface potential measurement technique was first developed. Both the effects of dust quantity and of the dust coverage pattern on the charging of a surface in a mesothermal plasma were analyzed. It was found that the dust layer acts as a capacitor and drives the potential of the surface beneath it more negatively. Additionally, patches of dust accumulation create plasma wake regions behind the dust patches. For a conducting surface, these plasma wake regions increase electron current collection, which drives the potential of the conducting plate further negative. An insulating surface, on the other hand, cannot distribute charge evenly and develops greater vi differential charging on its surface due to the combination of the dust layer capacitance and the plasma wake effects. It was found that the pattern of dust coverage on a surface affected the charging more than the quantity of dust coverage. I also validated the results with numerical modeling. To study photoelectron charging on the lunar dayside, two UV-C germicidal bulbs were used to generate ultraviolet radiation, which generated photoelectron current from the target surface. The results showed that as the dust coverage increased on a conducting plate, both the conducting plate and the dust-surface charged less positively. The same was true for an insulating material covered by dust. Additionally, the potential difference between the dust surface and the conducting or insulating surface was approximately the difference betweentheworkfunctionsofthetwomaterials. Becausethedifferenceintheworkfunctions between such materials was only a few eV’s, my results showed that differential charging was not a concern for a surface that was charged only by UV-illumination. Dusty surfaces exposed to other current sources in addition to photoemission, however, might exhibit more significant differential charging, and their geometry, orientation, and material properties must be considered. The potential threshold for electrostatic dischrage (ESD) to occur between a layer of lunar dust simulant, JSC-1A, and aluminum was also studied. To study dusty surface charging under the multibody-plasma interaction conditions, an experimental setup placing an “astronaut’s arm” in close proximity to both a conducting and a dusty surface in a mesothermal plasma was considered. This scenario is representative of an astronaut reaching over a surface to operate a piece of sensitive equipment or to pick up a tool. It was found that significantly greater differential charging occurred on an astronaut’s arm when it was in close proximity to the dusty surface. Since a dusty layer does not redistribute the charge stored to achieve a uniform surface potential, it increased the differential charging on the astronaut’s arm. More importantly, it was found that the electric field strength between the astronaut arm and both the conducting and insulating surfaces increased further downstream of the arm. Depending on the length of the astronaut’s arm and the distance between the arm and the surface, electrical breakdown or arcing could become a significant concern. TABLE OF CONTENTS vii Table of Contents Dedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv List of Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvi Chapter 1:Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1.1 Surface Charging in Plasma . . . . . . . . . . . . . . . . . . . . . . . 1 1.1.2 Spacecraft Charging . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.1.3 The Lunar Environment . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.1.3.1 The Lunar Plasma Environment and Lunar Surface Charging 7 1.1.3.2 Lunar Dust Environment . . . . . . . . . . . . . . . . . . . 11 1.1.4 The Asteroidal Environment . . . . . . . . . . . . . . . . . . . . . . . 14 1.1.5 Dusty Plasmas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 1.1.6 Dusty Plasma Literature Review . . . . . . . . . . . . . . . . . . . . 18 1.1.6.1 Charging of Dust Grains . . . . . . . . . . . . . . . . . . . . 18 1.1.6.2 Lunar Dust Grain Charging . . . . . . . . . . . . . . . . . . 22 1.1.6.3 Dust Grain Ion Trapping . . . . . . . . . . . . . . . . . . . 24 1.1.6.4 Dust Grain Photoemission . . . . . . . . . . . . . . . . . . . 26 1.1.6.5 Dust Grain Secondary Electron Emission . . . . . . . . . . . 27 1.1.6.6 Dust-Surface Interactions . . . . . . . . . . . . . . . . . . . 28 1.1.6.7 Dusty Spacecraft Charging . . . . . . . . . . . . . . . . . . 31 1.1.6.8 Electrostatic Discharge . . . . . . . . . . . . . . . . . . . . . 34 TABLE OF CONTENTS viii 1.2 Motivation and Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 1.3 Dissertation Outline and Approach . . . . . . . . . . . . . . . . . . . . . . . 36 Chapter 2:Experimental Setup and Surface Potential Diagnostics . . . . . . 38 2.1 Vacuum Facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 2.2 Electron Bombardment Gridded Ion Source . . . . . . . . . . . . . . . . . . 39 2.3 Ultraviolet Radiation Source . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 2.4 Plasma Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 2.4.1 Data Acquisition and Traversing System . . . . . . . . . . . . . . . . 41 2.4.2 Langmuir Probe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 2.4.3 Faraday Probe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 2.4.4 Emissive Probe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 2.4.5 Non-contacting Electrostatic Voltmeter . . . . . . . . . . . . . . . . . 45 2.5 Lunar Regolith Simulant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 2.6 Gore-Tex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 Chapter 3:DevelopmentandDemonstrationoftheDielectricSurfacePoten- tial Measurement Method . . . . . . . . . . . . . . . . . . . . . . . 47 3.1 Surface Potential Measurement Techniques and Difficulties . . . . . . . . . . 47 3.2 Embedded-Wire Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.3 Embedded-Wire Method Demonstration . . . . . . . . . . . . . . . . . . . . 52 3.3.1 Plasma Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.3.2 Dust-Covered Conducting Plate . . . . . . . . . . . . . . . . . . . . . 55 3.3.3 Dusty Surface Charging . . . . . . . . . . . . . . . . . . . . . . . . . 59 3.3.4 Dust Layer Capacitance . . . . . . . . . . . . . . . . . . . . . . . . . 61 3.4 Summary and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Chapter 4:Plasma Charging of Dusty Conducting Surface: Dust Coverage Effects on Spacecraft Charging . . . . . . . . . . . . . . . . . . . . 64 4.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 4.2 Plasma Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 4.3 Dusty-Plate Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 4.4 Current Collection-Capacitor Model . . . . . . . . . . . . . . . . . . . . . . . 67 4.5 Dusty-Surface Charging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 4.6 Simulation Study Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 4.6.1 Simulation Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 4.6.1.1 Governing Equations . . . . . . . . . . . . . . . . . . . . . . 70 4.6.1.2 Normalization . . . . . . . . . . . . . . . . . . . . . . . . . . 71 4.6.1.3 PIC Loop . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 TABLE OF CONTENTS ix 4.6.1.4 Surface Charging . . . . . . . . . . . . . . . . . . . . . . . . 73 4.6.2 Simulation Domain Setup . . . . . . . . . . . . . . . . . . . . . . . . 74 4.7 Simulation Study: Dust Pattern Effects . . . . . . . . . . . . . . . . . . . . . 75 4.8 Simulation Study: Dust Coverage vs. Dust Thickness . . . . . . . . . . . . . 79 4.9 Summary and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Chapter 5:Plasma Charging of Dusty Insulating Surface: Dust Coverage Effects on Space Suit Charging . . . . . . . . . . . . . . . . . . . . 81 5.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 5.2 Plasma Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 5.3 Dusty Gore-Tex Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 5.4 Dusty Gore-Tex Charging . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 5.5 Differential Charging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 5.6 Summary and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Chapter 6:Dusty Surface Charging Due to Ultraviolet Radiation . . . . . . 90 6.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 6.2 Photoelectron Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 6.3 Dusty-Surface Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 6.4 Photoelectron Charging of Dusty Conducting-Surface . . . . . . . . . . . . . 97 6.5 Photoelectron Charging of Dusty Insulating Surface . . . . . . . . . . . . . . 100 6.6 Electrostatic Potential Threshold Between JSC-1A and Aluminum . . . . . . 102 6.6.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 6.6.2 Dusty-Surface Charging with Negatively Biased Aluminum Plate . . . 104 6.6.3 Electrostatic Discharge Testing . . . . . . . . . . . . . . . . . . . . . 105 6.7 Summary and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Chapter 7:Plasma Charging of Dusty Surfaces During Multibody-Plasma Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 7.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 7.2 Plasma Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 7.3 Astronaut-Arm Charging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 7.4 Summary and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Chapter 8:Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . 116 8.1 Conclusions and Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . 116 8.1.1 “Dusty Surface" vs. “Dust-in-Plasma" . . . . . . . . . . . . . . . . . . 116 8.1.2 Conducting Surfaces: Clean vs. Dusty . . . . . . . . . . . . . . . . . 117 8.1.3 Insulating Surfaces: Clean vs. Dusty . . . . . . . . . . . . . . . . . . 118 TABLE OF CONTENTS x 8.1.4 Multibody-Plasma Interactions . . . . . . . . . . . . . . . . . . . . . 119 8.2 Recommended Future Research . . . . . . . . . . . . . . . . . . . . . . . . . 120 8.2.1 Expand Embedded-Wire Method . . . . . . . . . . . . . . . . . . . . 120 8.2.2 Deep Dielectric Charging . . . . . . . . . . . . . . . . . . . . . . . . . 120 8.2.3 Ion-Induced Secondary Electron Emission . . . . . . . . . . . . . . . 120 8.2.4 Dusty-Surface Charging in Plasma Wake Region . . . . . . . . . . . . 121 8.2.5 Charged Dust Motion . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 LIST OF FIGURES xi List of Figures 1.1 Schematic scenario of secondary electron suppression by potential wells during spacecraft charging in sunlight [1] . . . . . . . . . . . . . . . . . . . . . . . . 3 1.2 Emission of electrons and ions from a negatively charged surface to reduce negative charging [2] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.3 Images taken by NASA’s Lunar Reconnaissance Orbiter show some perma- nently shadowed regions near the lunar south pole [NASA] . . . . . . . . . . 7 1.4 Schematic of the lunar electrostatic environment in the solar wind [3] . . . . 8 1.5 Illustration of a local plasma void forming on the face opposite the solar wind flow [4] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.6 Sketch of the expected potential and field distribution when the Moon is in the solar wind [5] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1.7 Median electrostatic potential of the lunar surface in the solar wind, in spheri- calSelenocentricSolarEclipticcoordinates, relativetothelocalplasma. Black line indicates the average wake boundary at spacecraft altitude [6] . . . . . . 11 1.8 Glassy (left) and rocky (right) samples of lunar regolith [7] . . . . . . . . . . 12 1.9 Unprocessed images of horizon glow observed by Surveyor 5, 6, and 7 [8] . . 13 1.10 Sketches of sunrise with “horizon glow” and “streamers” viewed from lunar orbit by astronaut E.A. Cernan during the Apollo 17 mission [9] . . . . . . . 13 1.11 Number of events recorded by LEAM sensors [7] . . . . . . . . . . . . . . . . 13 1.12 Dust adhering to Apollo 17 astronaut Jack Schmitt’s spacesuit [10] . . . . . 14 1.13 Finger mark made by Apollo 12 astronaut Charles “Pete” Conrad Jr. in a layer of dust on the Surveyor 3 mirror [8] . . . . . . . . . . . . . . . . . . . . 14 1.14 Images of asteroid Itokawa and its regolith distributions [11] . . . . . . . . . 15 1.15 Dust-plasma interactions depending on inter-dust distance and Debye length 17 1.16 Langmuir probe characteristics obtained under identical conditions, except for the absence (upper plot) or presence (lower plot) of kaolin dust. In the lower characteristic, the dust dispenser is abruptly turned off near the end of the trace to check that the electron current returns to the no-dust value [12]. . . 19 1.17 Temporal evolution of charge number [13] . . . . . . . . . . . . . . . . . . . 21 1.18 Measured charges as function of the energy of the fast electron beam for glass microspheres (top panel) and lunar simulants JSC-1 and MLS-1 (bottom panel) [14] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 LIST OF FIGURES xii 1.19 Average of the measured charges for MLS-1, JSC-1, and Apollo-17 soil sam- ples [15] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 1.20 Trapped hydrogen ion orbiting around a negatively charged dust grain [16] . 25 1.21 Model geometry for secondary electron emission from a small spherical grain [17] 28 1.22 Potential contour plot (solid curves indicate negative potential, dashed curves indicate positive potential) illustrating strong potential well behind (to the right of) the grain [18] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 1.23 Schematic diagram of the experimental apparatus [19] . . . . . . . . . . . . . 30 1.24 (a) Experimental configuration for dust transport on a planar surface (b) potential distributions 1 mm above the surface (c) images showing dust trans- port into shadow region (right to left) with beam energy at 80 eV [20] . . . . 32 1.25 Thin stream of dust particles flow around a negative wire [21] . . . . . . . . 33 1.26 Lander close to a crater on the Moon simulated by SPIS [22] . . . . . . . . . 34 1.27 Simulation of one comet day of Philae. Before sunrise, 06:00 LT, (right) did not collect much dust. At 12:00 LT (left), about 0.0001% of the surface of Philae is covered [22] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2.1 Primary vacuum chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 2.2 Ion source and electrical configuration . . . . . . . . . . . . . . . . . . . . . . 39 2.3 A single UV-C germicidal bulb . . . . . . . . . . . . . . . . . . . . . . . . . . 40 2.4 Traversing system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 2.5 Langmuir probe and electrical configuration . . . . . . . . . . . . . . . . . . 42 2.6 Example of I-V curve generated by Langmuir probe data . . . . . . . . . . . 43 2.7 Faraday probe and electrical schematic . . . . . . . . . . . . . . . . . . . . . 43 2.8 Emissive probe and electrical schematic . . . . . . . . . . . . . . . . . . . . . 44 2.9 Trek ESVM non-contacting surface potential measurement method . . . . . 45 2.10 JSC-1A lunar regolith . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 3.1 Sample plate measurement setup with Trek non-contacting ESVM probe and measurement plate placed outside of the vacuum chamber . . . . . . . . . . 50 3.2 I-V trace of target aluminum plate . . . . . . . . . . . . . . . . . . . . . . . 51 3.3 Schematic showing the layout of the plasma source, sample plate, and the plasma flow field scan region . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.4 Photo of experimental setup inside the vacuum chamber . . . . . . . . . . . 53 3.5 Plasma environment above the sample plate. The centerline of the plasma beam is at z = 25.4 mm. The reference potential is the potential at x = 179 mm, z = 25.4 mm. The potential values plotted are offset by the reference potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 3.6 Schematic of diagnostic wire positions on the sample plate . . . . . . . . . . 56 3.7 Photo of a sample plate covered by a layer of JSC-1A regolith simulant . . . 56 LIST OF FIGURES xiii 3.8 Illustration of diagnostic wire height in experimental setup and dust layer thickness considered in experiment . . . . . . . . . . . . . . . . . . . . . . . 58 3.9 1-D potential profile of the ambient plasma and sample plate along the z- direction at x = 254.0 mm . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 3.10 Sample potential within dust layer measured by embedded wires . . . . . . . 60 4.1 A schematic showing the layout of the plasma source, sample plate, and the plasma flow field scan region . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 4.2 Plasma environment above the sample plate. The centerline of the plasma beam is at z = 25.4 mm. The reference potential is the potential at x = 179 mm, z = 25.4 mm. The potential values plotted are offset by the reference potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.3 Dusty plate sample setups . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 4.4 PIC loop performed at each timestep . . . . . . . . . . . . . . . . . . . . . . 72 4.5 Charge distribution and deposition onto mesh nodes [23] . . . . . . . . . . . 73 4.6 Simulation domain setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 4.7 Simulation dusty-sample potential field . . . . . . . . . . . . . . . . . . . . . 76 4.8 Simulation dusty-sample ion density . . . . . . . . . . . . . . . . . . . . . . . 77 4.9 Simulation dusty-sample electron density . . . . . . . . . . . . . . . . . . . . 78 4.10 Thick checkerboard simulation . . . . . . . . . . . . . . . . . . . . . . . . . . 79 4.11 Thick full dust coverage simulation . . . . . . . . . . . . . . . . . . . . . . . 79 5.1 Clean and dusty insulating surface flux . . . . . . . . . . . . . . . . . . . . . 82 5.2 Schematic showing the layout of the plasma source, sample plate, and the plasma flow field scan region . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 5.3 Plasma environment above the sample plate. The centerline of the plasma beam is at z = 19.1 mm. The reference potential is the potential at x = 179 mm, z = 19.1 mm. The potential values plotted are offset by the reference potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 5.4 Clean and dusty Gore-Tex sample setups . . . . . . . . . . . . . . . . . . . . 85 5.5 h1 and h2 plate setups for clean and dusty samples . . . . . . . . . . . . . . 86 5.6 Clean and dusty Gore-Tex surface potentials . . . . . . . . . . . . . . . . . . 87 6.1 Surface charging by UV illumination . . . . . . . . . . . . . . . . . . . . . . 91 6.2 A schematic showing the experimental setup of the UV source and target sample plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 6.3 Aluminum plate I-V curves as a function of cage bias . . . . . . . . . . . . . 93 6.4 Aluminum plate I-V curves with cage biased at -20 V and collecting plate biased at +30 V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 6.5 Dusty-sample setups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 6.6 Gore-Tex sample setups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 LIST OF FIGURES xiv 6.7 Dusty aluminum plate potential contours . . . . . . . . . . . . . . . . . . . . 99 6.8 Clean and dusty Gore-Tex potential contours . . . . . . . . . . . . . . . . . . 101 6.9 Electrostatic discharge dusty sample setup . . . . . . . . . . . . . . . . . . . 103 6.10 Electrostatic discharge test wiring setup . . . . . . . . . . . . . . . . . . . . 103 6.11 Example of electrostatic discharge event detected by oscilloscope . . . . . . . 104 7.1 Plasma flux to insulating surface near clean, conducting and dusty surface . 108 7.2 Schematicshowingthelayoutoftheplasmasource,astronautarm,anddusty/- conducting surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 7.3 Setup of astronaut arm near clean conducting and dusty surface . . . . . . . 109 7.4 Astronaut arm and dust layer embedded wire setup . . . . . . . . . . . . . . 110 7.5 Plasma environment above the sample plate. The centerline of the plasma beam is at z = 19.1 mm. The reference potential is the potential at x = 179 mm, z = 19.1 mm. The potential values plotted are offset by the reference potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 7.6 Plasma potential in Y-Z plane around the astronaut’s arm . . . . . . . . . . 111 7.7 Average, minimum, and maximum ambient plasma parameters at various x- positions along beam centerline . . . . . . . . . . . . . . . . . . . . . . . . . 113 LIST OF TABLES xv List of Tables 1.1 Physical comparison of the Moon and Earth [24] . . . . . . . . . . . . . . . . 6 1.2 Typical ambient plasma properties and resulting lunar surface potentials [25] 8 3.1 Floating potential measurements . . . . . . . . . . . . . . . . . . . . . . . . 52 3.2 Averageambientplasmaparametersatvariousx-positionsalongbeamcenterline 54 3.3 Wire tip height above the sample plate . . . . . . . . . . . . . . . . . . . . . 57 3.4 Aluminum plate floating potential and dust surface potential . . . . . . . . . 61 3.5 Total charge on dust layer and charge per dust grain . . . . . . . . . . . . . 62 4.1 Averageambientplasmaparametersatvariousx-positionsalongbeamcenterline 66 4.2 Dust coverage percentage and floating potential with respect to ambient plasma 69 4.3 Normalization reference parameters . . . . . . . . . . . . . . . . . . . . . . . 71 4.4 Normalized ion and electron velocities and temperature . . . . . . . . . . . . 75 4.5 “Thick" simulation results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 5.1 Averageambientplasmaparametersatvariousx-positionsalongbeamcenterline 83 5.2 Gore-Tex and dust surface potentials . . . . . . . . . . . . . . . . . . . . . . 87 6.1 Maximum photoelectron temperature . . . . . . . . . . . . . . . . . . . . . . 94 6.2 Ambient plasma parameters 0.8 cm above target sample surface determined by Langmuir probe sweeps . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 6.3 Average measured dusty sample surface potentials and charge density . . . . 98 6.4 Average measured Gore-Tex and dust layer surface potentials . . . . . . . . . 100 6.5 Average dust layer potentials with biased aluminum plate . . . . . . . . . . . 104 6.6 Approximate electric field strength as function of dust thickness . . . . . . . 105 7.1 Wire position distance from plasma source exit . . . . . . . . . . . . . . . . . 109 7.2 Averageambientplasmaparametersatvariousx-positionsalongbeamcenterline110 7.3 Average ambient plasma parameters at various x-positions along beam cen- terline of the three different plasma environments presented . . . . . . . . . . 112 7.4 Measured astronaut arm surface potentials with clean aluminum plate [V] . . 114 7.5 Measured astronaut arm surface potentials with dusty surface [V] . . . . . . 114 LIST OF SYMBOLS xvi List of Symbols C Capacitance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 s Surface potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Q Charge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 I j Current source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 I i Ion current . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 I e Electron current . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 I ph Photoelectron current . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 I sec Secondary electron current . . . . . . . . . . . . . . . . . . . . . . . . . 1 L 0 Characteristic length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 D Debye length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 0 Permittivity of free space . . . . . . . . . . . . . . . . . . . . . . . . . . 2 r d Dust radius . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 d d Inter-dust distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 C iso Capacitance of isolated dust grain . . . . . . . . . . . . . . . . . . . . . 17 rd Relative permittivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 n d Dust density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 n d0 Ambient dust density . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 V Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 E Photon energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 LIST OF SYMBOLS xvii h Planck’s constant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 c Speed of light . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Wavelength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 I e;sat Electron saturation current . . . . . . . . . . . . . . . . . . . . . . . . . 42 A probe Probe collecting surface area . . . . . . . . . . . . . . . . . . . . . . . . 42 0 Beam accelerating voltage . . . . . . . . . . . . . . . . . . . . . . . . . . 44 p Plasma potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 V Potential drop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 R Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 I Current . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 C s Ion acoustic velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 C parallel Parallel plate capacitance . . . . . . . . . . . . . . . . . . . . . . . . . . 61 A Surface area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 d Dust layer thickness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 plate Plate potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Q d Dust grain charge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Charge density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Q total Total charge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Q d;iso Isolated dust grain charge . . . . . . . . . . . . . . . . . . . . . . . . . . 62 cc Floating potential from current collection . . . . . . . . . . . . . . . . . 68 dust Dust square potentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 sheath Sheath potentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 s;est Estimated plate potential . . . . . . . . . . . . . . . . . . . . . . . . . . 68 E field Electric field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 A dust Dust square surface area . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 B Magnetic field vector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 LIST OF SYMBOLS xviii E Electric field vector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 Current density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 F Force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 q Electric charge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 f Floating potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 V b Bias voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 T ph Photoelectron temperature . . . . . . . . . . . . . . . . . . . . . . . . . 93 E Photon energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 W Work function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 I ph Photoelectron current . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 J ph Photoelectron current density . . . . . . . . . . . . . . . . . . . . . . . . 95 Y Photoelectric yield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 1 Chapter 1: Introduction 1.1 Background 1.1.1 Surface Charging in Plasma A surface immersed in a plasma will collect current and can be modeled by Equation 1.1, where C is capacitance of the object, s is surface potential, and Q is charge of the object. In a steady state, the surface will charge to a potential at which the total current collected by the surface will be equal to zero. This can be modeled as Equation 1.2, where I j are the different current sources collected by the surface. C d s dt = dQ dt (1.1) dQ dt = N X j=1 I j ( s ) = 0 (1.2) This dissertation considers surface charging in the solar wind plasma environment. For suchapplications,thecurrentsourcestothesurfaceinclude: 1)thesolarwind,whichcontains both electrons and ions and produces both negative and positive currents, respectively; 2) secondary electrons, which are generated by electron and ion impact and which result in an incidentpositivecurrent; and3)photoelectrons, whicharegeneratedbythecollectionofsolar photons and which result in an incident positive current [25]. The current balance condition is given by Eq. 1.3, whereI i is ion current ,I e is electron current,I ph is photoelectron current, and I sec is secondary electron current [26]. X k I k ( s ) =I i ( s ) +I e ( s ) +I ph ( s ) +I sec ( s ) = 0 (1.3) The magnitudes of the current collected are dependent on both the plasma sheath and the ambient plasma flow, and the plasma sheath thickness,d sh , is proportional to the plasma 1.1 Background 2 Debye length, D , (Eq. 1.4). For an object whose characteristic length, L 0 , is much greater than D , the current collection is also known as the space-charge-limited condition, and the sheath formed over the object is considered to be a thin sheath. In the space-charge-limited condition, all of the particles that enter the sheath are collected. WhenL 0 d sh , the sheath formed over the object is considered to be a thick sheath, and only particles within a certain impact parameter can be collected. This is also known as the orbital motion limited (OML) condition. D = r 0 kT e n e e 2 (1.4) The photoelectron current is dependent upon the flux of solar ultraviolet (UV) radiation onto the surface. On the sunlit hemisphere, the photoelectron current dominates the surface charging,anditchargessurfacesslightlypositively. Infact,asurfacewillchargetoapotential at which the emitted photoelectrons are drawn back to the surface, which is generally around thephotoelectrontemperature. Secondaryelectronemission,ontheotherhand,isconsidered to be negligible outside of the plasma sheet because particle energies are low, and it is highly dependent on the properties of the material [6]. 1.1.2 Spacecraft Charging Spacecraft charging is defined to be the charging of spacecraft surfaces or components to electricalpotentialsthataredifferentfromthepotentialofthesurroundingspaceplasma[27], and such charging follows the current balance equation, Eq. 1.3. Because spacecraft charging is highly dependent upon the properties of the material and because each material has unique properties (ie. secondary electron emission yield, photoemission yield, bulk electrical resistivity, and surface resistivity), the surfaces and components of a spacecraft can charge to different potentials. This condition is known as differential charging. Spacecraft can attain large negative potentials in eclipses, and they can attain slightly positive potentials ontheirsunlitareasduetodifferencesincurrentcollection. Hence, largepotentialdifferences can be created between its surfaces. Strong electric fields arise as a result of this differential 1.1 Background 3 Figure 1.1: Schematic scenario of secondary electron suppression by potential wells during spacecraft charging in sunlight [1] charging, which increases the risk of discharge and breakdown between the different surfaces. Therefore,spacecraftgeometryplaysaverysignificantroleinmitigatingdifferentialcharging. The holes in spacecraft surfaces, the different materials at each location along the spacecraft, and the complex geometry created by antennas and solar arrays will all have to be considered when determining how a given spacecraft will charge [28]. Figure 1.1 illustrates secondary electron suppression by potential wells for a spacecraft in sunlight. The spacecraft body charges positively due to UV-illumination, and secondary electrons are emitted by dielectric surfaces due to high energy ambient electrons. However, the secondary electrons are trapped by the positively charged spacecraft and are reabsorbed because their energies are only a few eV’s. The growing sophistication of spacecraft design has led to increased concern over space- craft environmental interactions that are associated with plasmas. Extensive research has been performed on spacecraft charging since the 1960’s. Accurate measurements of the charging on spacecraft and rockets have been made, and self-consistent charging models that include secondary emission and photoelectron current have been developed. The Space- craft Charging at High Altitudes (SCATHA) program was launched in 1979, and it deter- mined the response of the spacecraft to charging and evaluated the techniques to correct the 1.1 Background 4 problem [29]. Reviews by Garrett 1981 [30] and Whipple 1981 [31] summarized the major theoretical and observational findings. Discharge on spacecraft surfaces can disturb scientific measurements that are made onboard, can affect communications, control, and operations of spacecraft, and may be harmful to the health of the electronics within the spacecraft [2]. Many experiments have been conducted to determine how these materials charge in varying plasma environments. Paulmier et al., 2009 [32] used an electron beam gun to charge samples of Kapton and Teflon, two commonly-used materials in space, to determine how those materials would cope under electronirradationinspace. Theyfoundthatprolongedexposuretoelectronirradationcould lead to a reduction in the charging level for Kapton. On the other hand, they found that Teflon experienced enhanced electric surface potential and charging kinetics, which could lead to a high risk of discharge. Studies such as these will assist with the determination of how materials charge and how their properties may change over time due to long-term exposure. Internal charging has proven to be even more of a problem than spacecraft surface charg- ing due to electrostatic discharges that occur directly adjacent to victim circuits. Although the flux of high energy particles is generally low, the particles that penetrate the surface are trapped within the material, and charge accumulates. Secondary electrons cannot equili- brate this charging because they do not have enough energy to escape from the material. As a result, electrostatic discharge due to internal charging may cause serious problems due to its close proximity to sensitive circuits [28]. The problems of internal charging will only get worse in the future because economics is forcing the use of less and less shielding, of lighter weightstructuralmaterials, andeventheeliminationofFaradaycageconstructions[29]. Less shielding and thinner protective layers will both allow particles with even lower energies to penetrate the material and cause internal charging. Numericalmodelinghasimprovedtheunderstandingofspacecraftcharging, andadvance- ments in such modeling have allowed for the modeling of dynamic and geometrically complex situations [27]. Using numerical simulations, both equilibrium and transient potentials can 1.1 Background 5 be determined. This is crucial because the worst charging can occur when equilibrium has not yet been established, such as during the transition from eclipse to daylight. Figure 1.2: Emission of electrons and ions from a negatively charged surface to reduce negative charging [2] Several charge mitigation techniques have been developed to decrease the severity of spacecraft charging. The best spacecraft charging mitigation technique would be to prevent the buildup of high spacecraft surface potentials relative to the ambient plasma and relative to each other, but differing material properties and space weathering variation over time make doing so difficult. Therefore, active methods, such as emitting electrons from a hot filament to reduce negative potentials or using a low-energy ion beam to charge surfaces more positively, have been introduced. Passive methods of charge mitigation have also been developed, such as the use of sharp spikes, which protrude from charged surfaces to generate very large electric fields and emit electrons, reducing the negative potential of the conducting surface connected to the spike [2]. Figure 1.2 illustrates the emission of electrons and ions from a negatively charged surface to mitigate surface charging. The low-energy ions return to the negatively charged surface, reducing its charge, and generate secondary electrons, which further reduce the surface’s negative charge. 1.1 Background 6 1.1.3 The Lunar Environment The Moon is an alien environment compared to the Earth. Its gravitational force is six times weaker due to its smaller mass and radius; its surface is covered with several meters to tens of meters of loose regolith; and its lack of a global magnetic field and absence of a significant atmosphere expose the lunar surface to the surrounding solar wind and magnetospheric plasma, as well as solar energetic particles and micrometeroids. The Moon is located 384,400 km from the center of the Earth, and its radius is four times smaller than the Earth’s. Due to the close proximity and size of the Moon to the Earth, the Moon is tidally locked. This results in one hemisphere always facing the Earth, known as the “nearside,” and the other hemisphere always hidden from the Earth, known as the “farside.” Due to tidal locking, the lunar surface’s topography also creates permanently shadowed regions, as seen in Figure 1.3. Table 1.1: Physical comparison of the Moon and Earth [24] Property Moon Earth Mass 7:353 10 22 kg 5:967 10 24 kg Radius 1738 km 6371 km Surface area 37:9 10 6 km 2 510:1 10 6 km 2 Mean density 3:34 g/cm 3 5:517 g/cm 3 Gravity at equator 1:62 m/sec 2 9:81 m/sec 2 Escape velocity at equator 2:38 km/sec 11:2 km/sec Sidereal rotation time 27.322 days 23.9345 hr Mean surface temperatures 107 C day;173 C night 22 C Temperature extremes 233 C to 123 C 89 C to 58 C Atmosphere 10 4 molecules/cm 3 day 2:5 10 19 molecules/cm 3 2 10 5 molecules/cm 3 night Magnetic field 0 (small paleofield) 24-56 A/m TheMoonhasatenuousatmospherethatisapproximately14ordersofmagnitudesmaller than Earth’s atmosphere. It is composed of argon, helium, neon, sodium, potassium, and hydrogen, and is generated by both sputtering and outgassing. The daytime and nighttime gas concentrations in the atmosphere can vary by an order of magnitude due differences in the to solar heating of the lunar soil [24]. 1.1 Background 7 Figure 1.3: Images taken by NASA’s Lunar Reconnaissance Orbiter show some permanently shadowed regions near the lunar south pole [NASA] 1.1.3.1 The Lunar Plasma Environment and Lunar Surface Charging The lack of an atmosphere and of a global magnetic field exposes the lunar surface to both the space plasma and the solar radiation. The Moon’s position with respect to the Earth determines which plasma environment the lunar surface sees at any given time. When the Moon is in the solar wind, the lunar surface sees a relatively low temperature plasma with a flowing speed of 400 to 600 km/sec [24]. The solar wind particle density normally ranges from 1 to 20 particles per cm 3 . The solar wind is the main source of several of the elements in the lunar atmosphere and causes sputtering and ionization of the lunar surface, producing heavier pickup ions [6]. When the Moon is in the Earth’s magnetosphere, it is exposed to a very rarefied plasma in the magnetotail lobes and to more energetic and turbulent plasma in the plasma sheet and magnetosheath [25]. Table 1.2 summarizes the electron properties and lunar surface potentials measured by Lunar Prospector in the various environments. 1.1 Background 8 Table 1.2: Typical ambient plasma properties and resulting lunar surface potentials [25] Tail Lobe Plasma Sheet Solar Wind Wake Electron density 0:001 0:5 cm 3 0:01 1 cm 3 0:5 10 cm 3 0:001 0:1 cm 3 Electron temperature < 100 eV 100 eV to 2 keV 5 - 30 eV 50 - 150 eV Lunar surface potential -150 to 0 V -1000 to 0 V < 20 V -200 to 0 V The lunar terminator region, shown in Figure 1.4, is a particularly complex region for plasma interactions because the lunar surface transitions from sunlight-driven positive sur- face potentials to plasma-charged negative surface potentials [4]. Local wakes are also gen- erated when the horizontally-flowing supersonic solar wind flows across the surface of a mountain or spacecraft and expands vertically downward into the shadowed region located behind it [33], as illustrated in Figure 1.5. The local wake forms because a void is created, causing the more mobile electrons to expand and fill into that void. This creates a charge- separation electric field and a potential drop across the wake boundary. This potential drop slows the electron expansion and accelerates the ions along the field lines into the wake at a velocity proportional to the ion acoustic velocity [6]. Local near-surface electric fields normal to the surface form within the wake, and high temperature electrons can also be present, thereby creating a potentially-dangerous charging environment. Determining the plasma environment that is generated by a wake is critical to understanding local processes, such as surface charging, electrostatic dust transport, and space weathering. Figure 1.4: Schematic of the lunar electrostatic environment in the solar wind [3] 1.1 Background 9 Figure 1.5: Illustration of a local plasma void forming on the face opposite the solar wind flow [4] The lunar surface also encounters fluxes of solar energetic charged particle (SEP) events, such as coronal mass ejections, which are produced intermittently during periods of solar maximum. Solar flares can also emit and accelerate large fluxes of high energy electrons and energetic nuclei [24]. The Moon also absorbs solar electromagnetic radiation, such as ultraviolet, extreme ultraviolet, and X-ray. The radiation and particle energies that impinge on the lunar surface vary from solar wind ions at5 - 30 eV, to high-energy solar cosmic ray particles emitted intermittently at0.5 - 1 MeV, to high-energy galactic cosmic ray particles with energies greater than 1 GeV. The space plasma, energetic particle environments, and solar UV determine lunar surface charging. Manka 1973 [5] first solved for the surface potentials of the Moon as a function of the surface position in the solar wind, in the plasma sheet, and in high latitude regions of the geomagnetic tail. He determined that the photocurrent exceeded the solar wind plasma electron current on the sunlit side of the Moon and charged the surface to a few volts positive. As the surface transitioned from the terminator region to the nightside, the solar wind thermal electron current dominated and charged the surface to tens of volts negative. A sketch made by Manka of the expected potential and field distribution when the Moon is in the solar wind is shown in Figure 1.6. In the plasma sheet, where the electron temperature was much higher, he determined that the dayside charged between 2 to 20 V positive, and the nightside charged between 0 and -1800 V. Manka’s analytical approach was a starting 1.1 Background 10 point for determining the lunar surface potentials. However, it lacked accuracy and was over-simplified. Figure 1.6: Sketch of the expected potential and field distribution when the Moon is in the solar wind [5] The Apollo 14 and 15 missions deployed the Suprathermal Ion Detector Experiment (SIDE), which was used to measure and analyze positive ions near the lunar surface. Its results implied that sunlit surface potentials in the solar wind were on the order of tens of volts positive, and nightside surface potentials ranged from -50 to -100 V [34]. It is important to note, however, that secondary electron emission can complicate how surfaces charge on the night side. Depending on the incident electron energies, lunar secondary electron emission efficiencies can exceed unity, in which case a shadowed surface can charge positively [25]. The Apollo 14 mission also mobilized the Charged Particle Lunar Environment Experiment (CPLEE), an ion-spectrometer capable of measuring ion and electron energies between 40 eV and50keV.Itdeterminedthatinthelow-densitygeomagnetictaillobes, wherephotoelectron emission dominated, the surface potential charged as high as +200 V [35]. Most recently, the Lunar Prospector mission provided a better understanding of surface charging on the nightside. By measuring the potential drop between the spacecraft and the surface using electron reflectometry and modeling the spacecraft potential using a charging model specifically developed for the Lunar Prospector, scientists determined that the surface potentialreachedasnegativeas-100to-200Vonthenightsideandeitherapositivepotential or a small negative potential on the dayside. Data from the Lunar Prospector mission 1.1 Background 11 Figure 1.7: Median electrostatic potential of the lunar surface in the solar wind, in spherical Selenocentric Solar Ecliptic coordinates, relative to the local plasma. Black line indicates the average wake boundary at spacecraft altitude [6] produced the first global map of the average lunar surface potential in the solar wind and in the wake (relative to the local plasma just above the sheath) [6], shown in Figure 1.7. 1.1.3.2 Lunar Dust Environment The lunar surface is covered by a layer of regolith that is several meters thick, which lays on top of the primordial lunar bedrock. Due to the bombardment of about 10 6 kg of interplan- etary micrometeroids each year, the lunar surface is highly pulverized, and both micron and sub-micron sized secondary particles are ejected. These secondary particles form a perpetual dust cloud around the Moon or they blanket the lunar surface [7]. The lunar regolith has a mean grain size between 45 to 100 mm, and it is characterized to be sharp and glassy, analogous to fine-grained slag or terrestrial volcanic ash [24], as shown in Figure 1.8. Indi- vidual particles are primarily composed of agglutinates, which are aggregates of smaller soil particles bonded together by vesicular, flow-banded glass created by melting during microm- eteroid impacts [36]. Because lunar dust has a very low electrical conductivity and dielectric loss, electrostatic charging by the solar wind plasma, UV radiation, and high-energy particles does occur. 1.1 Background 12 Figure 1.8: Glassy (left) and rocky (right) samples of lunar regolith [7] The perpetual dust cloud around the Moon has been observed by the Galileo spacecraft, andthedust’selectrostaticpropertieshavecreatedsuchphenomenaaselectrostatictransport and dust scattering, as observed by the television cameras on Surveyors 5, 6, and 7. A distinct glow just above the lunar horizon (see Figure 1.9), referred to as horizon glow, was interpreted to be forward-scattered sunlight from a cloud of dust particles above the surface near the terminator [7]. The Apollo astronauts reported bright streamers high above the lunar surface during sunrise, as seen in astronaut E.A. Cernan’s sketch in Figure 1.10. The Lunar Ejecta and Meteorites (LEAM) experiment was deployed on the lunar surface by the Apollo 17 astronauts to characterize the lunar dust environment. Instead of the expected low impact rate from interplanetary and interstellar dust, LEAM registered hundreds of signals associated with the passage of the terminator, shown in Figure 1.11. It was suggested that the LEAM events were consistent with the sunrise/sunset-triggered levitation and transport of charged lunar dust particles [7]. In 2013, the Lunar Dust Experiment (LDEX), an in-situ dust detector onboard the Lunar Atmosphere and Dust Environment Explorer (LADEE) mission was launched. It was designed to characterize the variability of the dust in the lunar exosphere by mapping the size and spatial distributions of dust grains in the lunar environment as functions of local time and of the position of the Moon with respect to the magnetosphere of the Earth [37]. Based on the data obtained by LDEX, the dust ejecta cloud was found to be maintained by micrometeroid bombardment, as originally hypothesized. The density of the dust cloud rapidly increased towards the surface and showed strong temporal variability, most likely related to the stochastic nature of the meteroid impacts [37]. 1.1 Background 13 Figure 1.9: Unprocessed images of horizon glow observed by Surveyor 5, 6, and 7 [8] Figure 1.10: Sketches of sunrise with “horizon glow” and “streamers” viewed from lunar orbit by astronaut E.A. Cernan during the Apollo 17 mission [9] Figure 1.11: Number of events recorded by LEAM sensors [7] 1.1 Background 14 Figure 1.12: Dust adhering to Apollo 17 astro- naut Jack Schmitt’s spacesuit [10] Figure 1.13: Finger mark made by Apollo 12 astronaut Charles “Pete” Conrad Jr. in a layer of dust on the Surveyor 3 mirror [8] Fine grains from the lunar surface can be lofted above the surface due to human activities as well. The electrostatically-charged dust can be naturally transported under the influence of near-surface electric fields, resulting in dust transport and levitation at low altitudes [7]. The Apollo astronauts documented multiple issues that arose because of the dust, including dust that was adhering to their spacesuits after short extravehicular activities, as shown in Figure 1.12, as well as increased wear and tear, and obscured vision. This drastically reduced the lifetime of the spacesuits. Dust was also brought into the Lunar Module after moonwalks, making breathing without a helmet difficult and affecting vision in the cabin atmosphere [10]. It was also noted that layers of dust could accumulate on solar cells and thermal radiators, gradually reducing their effectiveness until they no longer functioned properly [38]. Figure 1.13 shows a finger mark made by Apollo astronaut Charles “Pete” Conrad Jr. in a dust layer that accumulated on a Surveyor 3 mirror. 1.1.4 The Asteroidal Environment Near-earthasteroids(NEA)aretypicallysmallerthanasteroidsinthemainbeltbecausesuch smaller asteroids are more easily dislodged from the main belt orbits. They are categorized into S-type, C-type, and M-type asteroids. S-type asteroids are the most common asteroids 1.1 Background 15 Figure 1.14: Images of asteroid Itokawa and its regolith distributions [11] in near-Earth space and are the parents bodies of the stony, ordinary chondrite meteorites. These asteroids are mainly composed of olivine and pyroxene and make up 15 to 20% of all known asteroids. Another 75% of all of the known asteroids are comprised of C-types, which dominate the asteroid belt and are likely to be the parent bodies of the primitive carbonaceouschondritemeteorites. C-typeasteroidshaveverylowalbedosandarecomposed of carbonaceous chondrite. The remainder of the known asteroids are mainly M-types, which have moderate albedo, as well as a flat, featureless spectrum at visible wavelengths, and are mainly metallic [39]. There is a variety of evidence that suggests that asteroids possess regoliths of unknown depths. Asteroid regoliths are thought to be dominated by grains coarser in size than the lunar regolith because asteroids most likely retain relatively less impact ejecta due to their weak gravitational fields. The Moon forms a locally-concentrated, size-sorted regolith due to repeated impacts, but impact ejecta on asteroids are ballistically spread over the entire surface, forming a generally uniform regolith [11]. Images from the Hayabusa mission of asteroid Itokawa have shown that there is a considerable amount of regolith distributed 1.1 Background 16 non-uniformly, but it was determined that vibrations due to impacts could trigger granular migrations and particle sorting. Figure 1.14 presents images of asteroid Itokawa and its regolith distributions. Due to reduced space weathering compared to the lunar surface, the dust grains range in size and distribution, and the chemical properties differ according to the type of asteroid. Hence, although there are few direction measurements of the asteroidal environment, one expects the plasma and dust environment at an asteroid to be somewhat similar to that at the lunar surface. The lack of an atmosphere and a magnetic field exposes an asteroid directly to the space plasma and to solar radiation. 1.1.5 Dusty Plasmas The focus of this dissertation is surface charging in a dusty plasma environment. To under- stand how a spacecraft charges in a “dusty” environment, it is important to first understand the electrostatic properties of dust grains. Dust particles coexist with electrons and ions in interplanetary space, with meteorites, with planetary rings, and with the lunar surface, just to name a few. They are immersed in plasma and illuminated by UV radiation. They build up electrostatic charges, and they respond to electromagnetic forces, in addition to gravita- tional force, drag, and radiation pressure. Charged dust particles can also alter the plasma environment they are within, and interactions between the dust and the plasma can lead to a variety of physical and dynamical consequences for both the dust and the plasma. Because dust particles are many orders of magnitude heavier than are other plasma particles, they can have many orders of magnitude larger time-dependent charges, which can influence the collective plasma behavior [40]. A single dust grain collects charge according probe theory and charges to a potential that is dependent on the ion and electron flux, on secondary electron emission, and on photoemission. Dust can also act as a source of electrons through thermionic emission, electric field emission, radioactivity, and exo-electron emission. Addi- tionally, the photoemission yield of fine-sized particles can be orders of magnitude greater than those of larger particles depending on particle size [41]. For typical space and laboratory plasmas, r d D ;d d , where r d is the dust radius and d d is the inter-dust distance. When D <d d , the dust is considered a collection of isolated 1.1 Background 17 (a) Dust-in-plasma (b) Dusty plasma (c) Dusty surface Figure 1.15: Dust-plasma interactions depending on inter-dust distance and Debye length screened grains, also known as “dust-in-plasma” (Fig. 1.15(a)). When D > d d , the dust particles behave as a collective ensemble, also known as a “dusty plasma” (Fig. 1.15(b)) [42]. The difference between “dust-in-plasma” and a “dusty plasma” is of significance. “Dust-in- plasma” particles are isolated, sphericaly-shaped dust grains, and they behave much in the same way as does a probe. They collect electrons and ions from the background plasma according to orbital-motion-limited (OML) theory, experience conventional Debye screening by electrons and ions, and their capacitance can be modeled as in Eq. 1.5, where rd is the relative permittivity [43]. The basic equations for isolated dust grain charging at steady state were shown in Eq. 1.1 and 1.2. C iso = 4 0 rd r d (1.5) For a “dusty plasma,” where dust grains are densely packed, the grains are electrically coupled with the neighboring grains because the inter-dust distance is much less than the Debye length. When a collection of dust grains whose inter-dust distance is less than the Debye length rests on a surface (Fig. 1.15(c)), a single sheath is formed over the dust sur- face, and the condition is known as a “dusty surface.” The “dusty surface” condition is what is studied in this dissertation. The grains in a dusty plasma will acquire a smaller elec- tron charge due to the depletion of electrons required by the equilibrium quasi-neutrality condition, given in Eq. 1.6, where n d is the dust number density [44]. en i0 =en e0 Qn d0 (1.6) 1.1 Background 18 Dust charging is also reduced when the ratio between the intergrain spacing and the plasma Debye radius is less than unity. In plasmas with dust grains of different sizes and shapes, it is likely that particles will charge differently [44]. 1.1.6 Dusty Plasma Literature Review 1.1.6.1 Charging of Dust Grains Whipple et al., 1985 [45] derived theoretically the potential distribution in a plasma con- taining dust grains where the Debye length can be either larger or smaller than the average intergrain spacing. In particular, they looked at the capacitance and charge of an individual dust grain in the presence of the neighboring grains and the surrounding plasma, and they found that the capacitance of a dust grain increases with the proximity of neighboring grains. When the ion density was significantly larger than the grain concentration, the actual electric shielding distance was approximately the average intergrain distance rather than the Debye shielding distance. This is the basis of the spherical capacitor model. The dust grain can be modeled as a spherical capacitor (Eq. 1.5), and the current collected is a function of the dust grain capacitance and surface potential (Eq. 1.1). Because the amount of charge stored in a capacitor is modeled by Eq. 1.7, where V is voltage, as the capacitance increases, the grain charge decreases to charge to the same potential. It is also noted that these conclusions may not be valid for very densely packed grains. Q =CV (1.7) Goree 1994 [46] reviewed and modeled dust particle charging in a plasma as well. He demonstrated that the charge on dust is reduced at high dust densities due to electron deple- tion on the particle, and he determined that the OML theory used to estimate isolated grain charging could be improved by including charge reduction at high dust densities, electron emission, and ion trapping. 1.1 Background 19 Figure 1.16: Langmuir probe characteristics obtained under identical conditions, except for the absence (upper plot) or presence (lower plot) of kaolin dust. In the lower characteristic, the dust dispenser is abruptly turned off near the end of the trace to check that the electron current returns to the no-dust value [12]. Barkan et al., 1994 [12] demonstrated reduced grain charge for “closely packed” grains (d < D ) experimentally. They generated a fully-ionized, magnetized Potassium plasma column and dispensed hydrated aluminum silicate dust grains of various sizes and shapes into the plasma. 90% of the dust grains were in the 1 - 15 mm range. Using a Langmuir probe, Barkan et al. found that the electron saturation current was smaller when dust was present in the plasma, shown in Figure 1.16. This was due to the fact that electrons were attached to dust grains of extremely low mobility and were not collected by the probe. Hazelton et al., 1994 [47] also determined the charging characteristics of dust grains in a plasma to validate the charging models that had been developed at the time. A beam of bismuth dust particles was generated by an oven and was then traversed into the main cham- ber, where the dust particles interacted with an argon plasma produced by a hot filament cathode discharge and with background argon gas. A small retarding potential analyzer (RPA) measured the electron temperature and density, and a tungsten probe measured the 1.1 Background 20 floating potential. The beam then entered a large RPA, which detected the charge on the dust particles. The results showed that increasing the plasma density increased the total cur- rent that was collected, and the drop in current collected was much faster than the increase in the retarding voltage magnitude. These features were consistent with increased charge accumulation on each grain as it passed through the plasma. Since there was more charge per grain, the total current carried by the dust beam increased. Because the charge to mass ratio of each grain was greater, each particle was now stopped by a larger retarding voltage. They found that when the dust grains were near equilibrium, the charge distribution was reasonably well described by the spherical capacitor charging model. Cui et al., 1994 [13] presented a numerical simulation on the fluctuations of the charge on a dust grain in a plasma. They claimed that charging processes commonly use a “continuous charging model,” which assumes that the charge on the grains is determined by current collection; however, due to the descrete nature of ions and electrons, Cui et al. argued that the charge on the dust grains should fluctuate. These fluctuations could then alter the motions of the dust in the plasma. The simulation results showed that the continuous model produced a smooth curve, and the discrete model fluctuated in discrete steps about the smooth curve, as is shown in Figure 1.17. They determined that charge fluctuated much more for small grains; on the other hand, the larger grains charged much more quickly but they did not fluctuate as much. Walch et al., 1995 [48] measured the charge on small grains of glass microballoons (rep- resenting a substance with a high secondary emission coefficient), graphite (low secondary emission coefficient), copper (metallic conductor), and silicon (important in plasma process- ing) in a plasma device containing both thermal electrons with energies of a few eV and monoenergetic suprathermal electrons. The dust grains ranged from 35 5 to 115 10 mm, and individual grains were dropped through an argon plasma and into a Faraday cup below the plasma. The Faraday cup was connected to a sensitive amplifier, and the height of the output pulse indicated the grain charge. The experimental results showed that the 1.1 Background 21 Figure 1.17: Temporal evolution of charge number [13] grains charged linearly with electron energies and followed the grain capacitance and charg- ing model. When electron energies were high enough to produce secondary electron emission, the grain potentials charged more positively. Zobnin et al., 2000 [49] performed a self-consistent molecular-dynamics numerical sim- lation on the charging of micron-sized particles in a low-pressure gas-discharge plasma, and they found that ion-neutral collisions, which produced charge-exchange ions, affected the potential of micron-sized dust particles at pressures on the order of several Pascals, which corresponded with ion mean-free-paths on the order of several millimeters. They showed that these ion-neutral collisions could lead to large errors when the OML theory was used to approximate the charge of dust particles levitating in a gas-discharge plasma. Lampe 2001 [50] analytically determined, though, that in a stationary plasma, and in the limit of small grains, where r d = D ! 0, OML theory was exactly correct. He demonstrated that for essentially all conditions of interest in connection with dusty plasma, the grain size was small enough to justify the use of OML theory for quantitative calculations of the ion cur- rent to the grain and thus of the grain charge and floating potential. It was also noted that potential barriers could have a more significant effect on shielding at ranges on the order of D , especially for large grains, small values of T i =T e , and high plasma density. 1.1 Background 22 1.1.6.2 Lunar Dust Grain Charging To understand how dust grains on the lunar surface charge, a simulated lunar regolith, such as Johnson Space Center One (JSC-1), Johnson Space Center 1A (JSC-1A), or Minnesota Lunar Simulant 1 (MLS-1), was often used. Horanyi et al., 1995 [14] measured the charge on grains of simulated lunar regoliths, MLS- 1 and JSC-1, in an argon plasma containing thermal electrons with energies of a few eV and a second population of nearly monoenergetic fast electrons with energies from 15 to 100 eV. The dust grains were dropped into the chamber from the top of the chamber and collected in a Faraday cup mounted on a diagnostic arm below the chamber. The Faraday cup was connected to a sensitive electrometer for measuring the charge on the grains. Experimental results showed that in the absence of secondary electron production (below 50 eV), the dust grainschargedlinearlywiththeelectronenergy. Forhigherbeamenergies,secondaryelectron production became significant, and the linear trend was broken, as seen in Figure 1.18. Walch et al., 1998 [15] also reported on an experimental study on the electrostatic charg- ing properties of MLS-1 and JSC-1. Similarly to the study by Horanyi et al., 1995, the dust grains were dropped from the top of the chamber and then collected in a Faraday cup mounted below. The dust grains were exposed to a beam of fast electrons with energies in the range of 20E 90 eV. Using dust grains with radii of 29 3 mm, experimental data showed that the dust grains charged linearly when exposed to beam energies below 50 eV, illustrated in Figure 1.19. At higher beam energies, secondary electron production became significant and broke the linear trend. The secondary electron yield for an Apollo-17 dust sample was also measured, and its values were quite similar to those of the MLS-1 and JSC-1 simulants for energies of 20E 90 eV. Sternovsky et al., 2002 [51] reported on the results of a simple experiment that investi- gated the contact charging properties of lunar regolith simulant, JSC-1, and Martian regolith simulant, JSC-Martian-1. When two materials were brought into contact, charge transfer could occur, and after the contact was broken, a non-zero charge remained on each species. This is known as contact charging. Electrification due to dust-dust collisions was thought 1.1 Background 23 Figure 1.18: Measured charges as function of the energy of the fast electron beam for glass microspheres (top panel) and lunar simulants JSC-1 and MLS-1 (bottom panel) [14] to be responsible for cohesive and adhesive forces between the dust grains in the lunar soil. In the experiment, dust was placed on a vibrating disc, which charged the grains by contact charging and dropped the dust particles into a Faraday cup below. The intergrain distance was kept larger than the grain size so that grains charged through contact with the dropper disc rather than through contact with one another. It was found that contact charge on the dust grain increased during the course of the experiment. Dust particles that dropped later into the experiment had, on average, more charge than earlier dust particles due to repeated contact between the dust particles and the disc surface. Contact charging as a function of dust size was also investigated, but the dust size range was too small to deter- mine any dependencies, though it was found that as size increased, the measured charge increased as well. It was found that charge measured on the grains due to contact charging could exceed the charge a dust particle typically collected in a low temperature space plasma 1.1 Background 24 Figure 1.19: Average of the measured charges for MLS-1, JSC-1, and Apollo-17 soil sam- ples [15] environment, depending on the work function of the surface, the dust size, and the degree of dust migration upon the surface. These results suggested that grains that were lifted off airless, planetary surfaces could carry a significant charge, regardless of the ambient plasma conditions, and they could be transported by strong local electric fields. 1.1.6.3 Dust Grain Ion Trapping Goree 1992 [16] presented a theory of ion trapping by a charged dust grain in a plasma, as is shown in Figure 1.20. Using the standard charging model, Eq. 1.1 and 1.5, Goree theorized 1.1 Background 25 Figure 1.20: Trapped hydrogen ion orbiting around a negatively charged dust grain [16] that negatively charged grains could hold or confine positive ions, which, in turn, could shield the grain from external fields. The ion shield density was determined by the incident ion flux, the capture cross section, and the confinement time. He also determined that the number of collisionally trapped ions was independent of the collisional mean free path. Using simulations, Goree found that ion trapping was negligible in space physics problems, where n i 10 12 m 3 , but trapping in laboratory plasmas, where n i 10 12 m 3 , was so strong that it was often in the saturation regime. This meant that previous calculations of electric forces acting on grains in laboratory plasmas that ignored trapping were probably invalid unless the calculations were performed in regions within a Debye length of a sheath. Lampe et al., 2001 [52] introduced a fully analytical method for calculating the distribu- tion of both trapped and untrapped ions. Considering a single stationary grain immersed in a non-flowing plasma consisting of singly charged positive ions and neutral molecules, they showed that the trapped ions dominated the shielding around the charged grain, due to ion-neutral charge-exchange collisions, and they confirmed Goree 1992’s findings that the numberofcollisionallytrappedionswasindependentofthecollisionalmeanfreepath. Under typical conditions, the inner part of the shielding cloud is made up primarily of trapped ions, 1.1 Background 26 and the potential, as a function of distance from the grain, is different from the results of the collisionless theory, which ignore ion and electron collisions. They concluded that the presence of a large population of trapped ions could profoundly change the interaction of a grain with other grains and with external forces. 1.1.6.4 Dust Grain Photoemission Feuerbacher et al. 1972 [53] performed experiments on the photoelectric properties of lunar surface fines number 14259,116 and found the work function (5 eV), photoelectric yield (7% at 900 Å), and energy distribution (peaks around 2 to 3 eV) of the photoemitted electrons using photons with energies from 4 to 21 eV. Abbas et al. 2002 [54] performed photoemission experiments by levitating 2-10 m diameter silica and polystyrene dust particles and illuminated them with UV radiation with a spectral resolution of 8 nm. They found that the energies of the emitted photoelectrons were much smaller than the assumed value of 1 eV, and the photoemission efficiency and yield were the same for negatively biased particles. This was because the biased surface potentials allowed photoelectrons to escape the particle. Abbas et al. 2006 [55] measured the photoelectric efficiences and yields of individual sub-micron and micron sized dust grains selected from sample returns of Apollo 17, Luna-24 missions, and JSC-1 simulant. The dust particles were illuminated by UV radiation with wavelengths of 120, 140, and 160 nm, and the photoelectric efficiencies of the dust grains were measured by direct measurement of the photoelectric yield and absorption efficiencies calculated using Mie theory. Their results indicated that photoelectric yields were dependent on the grain size. The yield increased by an order of magnitude when going from grains of sub-micron size to grains of several microns in size, and eventually, their yield reached an asymptotic value. Champlain et al. 2015 [56] studied the near-surface lunar dust conditions and found the upper (10 4 kg=s 2 for particles between 1 and 100 m sizes) and lower (10 6 kg=s 2 ) limits of the cohesive force between dusts using experimental and numerical simulations of dust chargingunderUVirradiationinthepresenceofanelectricfield. Theirresultssuggestedthat 1.1 Background 27 dust ejection by electrostatic forces was made possible by microscopic-scale amplifications due to soil irregularities. Nouzak et al., 2016 [57] presented laboratory measurements on the work function of a single micrometer-sized lunar dust simulant particle captured in an electrodynamic trap and irradiated by UV photons. Results showed that the work function of Minnesota Lunar Simulant was larger than 5 eV, and the photoelectron yield for photons of20 eV was around 0.05. 1.1.6.5 Dust Grain Secondary Electron Emission Chow et al., 1993 [17] discussed the role of secondary electron emission with a special empha- sisontheroleofgrainsize. Ithadpreviouslybeenunderstoodthatsecondaryemissionplayed an important role in determining the equilibrium grain potential, but the importance of the grain size on the secondary emission was not yet widely appreciated. Using both conducting and insulating spherical grains of various sizes, Chow et al. derived a model that showed that secondary emission yield increased with decreasing size and became very large for grains whose dimensions were comparable to the primary electron penetration depth, as is illus- trated in Figure 1.21. The secondary emission yield was found to be larger for insulators in comparison with conductors. It was also shown that the secondary electron emission from small spherical grains could lead to highly positive equilibrium surface potentials, and, inter- estingly, grains of different sizes could have opposite charges depending upon the thermal energies. Venturini et al., 1998 [58] experimentally studied the interaction of micron-sized NaCl particles with plasmas and primarily focused on their secondary electron emission. NaCl particles were suspended in the air using electro-dynamic suspension and subjected to a 500 eV electron beam for 30 minutes generated by a variable energy electron gun. A Faraday cup was located below the balance to record the total beam flux. Using the charge, size, and mass of the particle, the secondary electron emission could be calculated by knowing the net current to and from the particle. Preliminary results found a secondary electron yield of 1.06. 1.1 Background 28 Figure1.21: Modelgeometryforsecondaryelectronemissionfromasmallsphericalgrain[17] 1.1.6.6 Dust-Surface Interactions Murphy et al., 1991 [59] investigated the mutual interaction of plasmas and dust in the presence of a large material body. By assuming a low dust density and a random, Maxwellian velocity distribution, they found that the magnitudes of both the charge on the body and its potential were reduced by the presence of dust. Larger dust densities resulted in smaller body charges and potentials because the charged dust particles attracted ions toward and repelled electrons away from the body. Dust particles with a larger mean velocity were also more effective in reducing the spacecraft charge because they were not as affected by the body’s potential and electric field. Lampe et al., 2000 [18] looked into interactions between dust grains in a dusty plasma. In a stationary plasma, the electrostatic potential of a single grain could be calculated by OML theory. If the plasma was drifting, however, each grain generated a wake field potential, shown in Figure 1.22. Because the interaction between grains was highly anisotropic, the wake of one grain could have stable points that attracted other grains, which could result in crystalline dust structures in strongly-coupled dusty plasma. Lampe et al. also found that as long as the particle separation was large compared to the grain radius, the floating 1.1 Background 29 Figure 1.22: Potential contour plot (solid curves indicate negative potential, dashed curves indicate positive potential) illustrating strong potential well behind (to the right of) the grain [18] potential of the grains was not significantly modified by the presence of the other grains. On the other hand, when d < D , a reduced charge on each grain was sufficient to produce a more negative potential. Wang et al., 2007 [19] reported on their experimental investigation on the charging of dust particles resting on conducting and insulating surfaces beneath a plasma to give insight into the behavior of dust on planetary surfaces exposed to the solar wind plasma. The initial charge of the particles resting on the surface was an uncertainty and affected the initial electric force they experienced if and when they left the surface. It was also known that an isolated grain that is levitating or falling through a plasma charges to a floating potential that is likely to be different from the floating potential for a planar surface. An electrically insulating dust layer on a spacecraft that is sufficiently thick to cover the underlying conduc- tor, on the other hand, will charge to the floating potential for a flat suface if the dimensions of the layer are greater than the Debye length. Charging of a single insulating dust grain on a conducting surface is dependent upon the potential of the surface; a surface at the floating potential receives equal ion and electron flux, whereas a surface at a potential different from the floating potential will receive unequal fluxes. Wang et al. placed conducting (Ni) and 1.1 Background 30 Figure 1.23: Schematic diagram of the experimental apparatus [19] insulating (SiO 2 , Al 2 O 3 , and JSC-1 lunar regolith simulant) dust particles on a conducting surface biased above its floating potential to create an electron or ion dominant flux. Their experimental setup is shown in Figure 1.23. The results showed that the charge on the insu- lating dust particles saturated at a positive value for a plate that was biased more negatively than its floating potential, and the charge also saturated at a negative value when the plate was biased more positively than its floating potential. This demonstrated that the sign of the charge on the insulating dust was determined by the potential of the surface relative to the floating potential. Because there is currently no theory for the charging of dust resting on a surface in plasma, however, the results could not be compared with calculated values. Wang et al., 2011 [20] also investigated the dynamics of dust particles on a conducting surfaceinalaboratoryplasmatounderstandthephysicsofelectrostaticlunardusttransport. JSC-Mars-1dustsimulantpileswereplacednearthecenterofagraphiteplatebiasedto-80V with respect to the floating potential, and an argon plasma was created by electrons emitted from a hot filament. Results showed that over time, the dust began to spread uniformly into a ring formation. It was determined that the dust particles charged positively due to the biased graphite plate, resulting in electrostatic forces pointing outwards in the radial direction and upward in the vertical direction. This caused the dust particles to spread and 1.1 Background 31 hop on the graphite surface. This condition could be similar near the ion wake boundary on the nightside of the lunar surface when the Moon enters the Earth’s plasma sheet. Wang et al., 2011 [60] additionally demonstrated dust transport due to the generation of strong electric fields. In the experiment, electron beams up to 80 eV were used to bombard a pile of JSC-Mars 1 simulant, and a shadow region was created next to the dust pile with an insulating sheet, as shown in Figure 1.24. Large potential differences were created near the electronbeamimpact/shadowboundariesinthelaboratory, whichproducedsufficientlylarge electric fields to transport the negatively charged dust particles from the beam-illuminated region into the shadow. This motion could explain dust transport near boundaries created by high energy electron fluxes incident upon mountains and boulders on the lunar surface. Sheridan 2013 [61] used a stochastic model for the charging of a dust grain sitting on a surface exposed to plasma and solved for the time dependence of the average and variance of the charge fluctuations using a Monte Carlo simulation. He found that the fluctuations had zero mean, and the variance approached a constant value. His results predicted a decrease in charging time with plasma density. Meyer et al., 2014 [21] produced a dusty plasma using a DC glow argon discharge and captured single frame video images of dusty plasma flowing around a conducting wire, shown in Figure 1.25. The captured images illustrated the collective behavior of negatively charged dust particles in response to the background plasma, electric fields, and negatively biased obstacles. The images were obtained by laser illumination of the dust suspension with the reflected light recorded using a high-speed CMOS camera. 1.1.6.7 Dusty Spacecraft Charging Wang et al., 2008 [62] presented a particle simulation model for spacecraft charging and charged dust particle interactions on the lunar surface. Using average solar wind plasma conditions, they created a full particle particle-in-cell (PIC) simulation that simulated the plasma sheath. In particular, they looked at how the lunar surface and the spacecraft surface charged at multiple sun elevation angles and found that on average, the floating potential of the lunar surface ranged from -22 V at 0 to slightly positive at 10 . Wang et al. also 1.1 Background 32 Figure 1.24: (a) Experimental configuration for dust transport on a planar surface (b) poten- tial distributions 1 mm above the surface (c) images showing dust transport into shadow region (right to left) with beam energy at 80 eV [20] considered electrostatic levitation of dust particles and simulated how charged dust particles moved around the lunar lander. They found that dust at the lunar terminator region could become easily levitated due to lunar surface charging and could become deposited on the lander surface. Hsuetal., 2013[63][64]focusedonspacecraftcharginginthehighdustdensityregionnear Enceladus in Saturn’s E-ring. In a dust-rich environment, spacecraft charging is modified by 1.1 Background 33 Figure 1.25: Thin stream of dust particles flow around a negative wire [21] the complex plasma conditions as well as the dust-associated charging currents. Hsu et al.’s numerical investigation found that the spacecraft potential in a “dusty” environment could not be modeled precisely without knowing the spacecraft’s speed and the plasma and dust properties. They stated that dust size distribution variability was difficult to constrain, and there was still uncertainty regarding dust size and charging. From their models, though, they determined that a spacecraft charged less negatively when traversing through dust- rich regions. Their simulation results will need to be verified by Cassini dust and plasma measurements. Hessetal., 2015[22]furtherdevelopedtheSpacecraft-PlasmaInteractionSoftware(SPIS) to include the modeling of lunar dust charging, ejection, dynamics in the plasma, and depo- sition on surfaces, creating SPIS-DUST. SPIS is an open source tool for the simulation of the charging of materials in a plasma environment. Hess et al.’s developments allowed for dust deposition on spacecraft surfaces, as shown in Figure 1.26, and the deposit rate, dust layer width, and fraction of the surface covered could all be monitored. Using their simulation, they were able to model a lunar lander on a crater rim, as well as the Philae lander on 67P/Churyumov-Gerasimenko, as shown in Figure 1.27. 1.1 Background 34 Figure 1.26: Lander close to a crater on the Moon simulated by SPIS [22] Figure 1.27: Simulation of one comet day of Philae. Before sunrise, 06:00 LT, (right) did not collect much dust. At 12:00 LT (left), about 0.0001% of the surface of Philae is covered [22] 1.1.6.8 Electrostatic Discharge Electrostatic discharge has been a subject of extensive investigations, particularly on solar arrays. The highest power-generation voltage in orbit is 160 V on the International Space Station (ISS), but there is growing need for high power in future space missions, which requirehigh-voltagepowergenerationandtransmissiontominimizeenergylossduringpower transmission [65]. Higher voltages from 300 to 400 V, in particular, are of interest. 1.2 Motivation and Objectives 35 Cho et al., 2000 [66] found that arcing could occur at the triple junction of solar arrays, where the differential voltage between the insulator and conductor was 200 V or larger in low earth orbit (LEO). Cho et al., 2002 [67] also found that the current paths and amount of charge released through arcs was dependent on how far apart the arc point and the insulator were separated. Toyoda et al., 2004 [68] and Kawasaki et al., 2005 [69] found that the charging of satellites in sunlight was dominated by photoelectrons, but when part of an insulator surface was in a shadowed region, they found that large potential differences could form between the satellite body and insulator surface. In particular, they found that arcing occurred when the differential potential was greater than 400 V. Hosoda et al., 2006 [65] developed solar array technology capable of generating power at 400 V in a LEO plasma environment and found that covering surfaces with ethyltetrafluo- roethylene (ETFE) film suppressed arc inception by up to 800 V. Kawasaki et al., 2006 [70] investigated arc plasma propogation phenomena in a geo- stationary orbit (GEO) plasma environment and found that a neuralization current was provided by the charge stored on coverglasses and insulator substrates. They also found that less charge was neutralized as the distance from the arc position increased. Cho et al., 2013 [71] recently developed HORYU-II and achieved the highest photovoltaic powergenerationinorbitat350Vusingseries-connectedmicrosolarcells, whichwassuitable for other scientific experiments requiring high voltages, such as electric propulsion. Toyoda et al., 2015 [72] found that flashover was formed by a plasma bubble that expanded over the insulator surface and fed electrostatic energy stored on the insulator into the primary arc plasma. This could propagate as far as 3.4 m on a 4.8 m solar array with an average speed of 12 km/s after ESD inception. 1.2 Motivation and Objectives Based on the work that has been performed on dust grain charging, it is clear that no studies have looked into the charging of dusty surfaces in a mesothermal plasma. The goal 1.3 Dissertation Outline and Approach 36 of this dissertation is to study the effects of dust on plasma charging of both conducting and insulating surfaces through experimental, analytical, and numerical investigations. The results will be applicable to a dust covered surface at the lunar terminator region under the average solar wind plasma condition. Specifically, this dissertation investigates: 1. charging of a “dusty surface” vs. charging of “dust-in-plasma” 2. charging of a clean conducting surface vs. charging of a dusty conducting surface 3. charging of a clean insulating surface vs. charging of a dusty insulating surface Previous studies on dust charging have mostly considered the charging of single, isolated dust grains. However, results from such single-dust based measurements or models are only valid for the “dust-in-plasma" condition, where the inter-dust distance is much larger than the sheath thickness of each individual dust grain, and the dust charging is not affected by the presence of neighboring dust grains. Hence, my first objective was to measure the charge of dust grains for the “dusty surface" condition, where the inter-grain distance is much smaller than the Debye length and dust grains in the dust layer are no longer electrically isolated from neighboring grains, and to compare it with the charge for an isolated dust grain. Previous studies on spacecraft charging have focused on the charging of a “clean” con- ducting or dielectric surface in plasma, but when a spacecraft or astronaut is in a dusty environment, such as that found on the lunar surface and near asteroids, their surfaces will be covered by layers of dust particles. In this situation, the floating potentials of the surfaces will be determined by both the current collection condition and the charging properties of the dust layers. The second and third objectives were to determine the effects of dust on the charging of both conducting and insulating materials. 1.3 Dissertation Outline and Approach The following outline states the remaining contents of the disseration: 1.3 Dissertation Outline and Approach 37 InChapter2,theexperimentalfacilityisintroduced,thediagnosticsutilizedtomeasure plasma parameters are discussed, and the various sample materials are examined. In Chapter 3, a novel dielectric surface potential measurement method is presented and used to measure “dusty surface” charging. In Chapter 4, the effects of dust coverage on the charging of a conducting plate in a mesothermal plasma are presented, and numerical simulation results are compared. In Chapter 5, the effects of dust coverage on the charging of an insulating material in a mesothermal plasma are presented. In Chapter 6, results from the charging of both a dusty conducting surface and dusty insulating surface illuminated by UV-radiation are presented, and the determination of the electrostatic discharge potential threshold on dusty surfaces is performed. In Chapter 7, a study on the effects of dust coverage on plasma charging under multibody-plasma interaction conditions is presented. In Chapter 8, conclusions and suggestions for future work are given. Compared with previous work, which focused on single, isolated dust grain charging in plasma, thisdissertationinvestigatedthechargingofadustysurfaceinplasma. Additionally, previous work only considered the charging of dust grains in a stationary, thermal plasma, whereas results presented in this dissertation were found in a mesothermal plasma, similar to that found at the lunar terminator, and in a photoelectron dominant plasma, similar to that found on the lunar dayside. A novel method to directly measure the potential of dielectric and insulating materials in plasma without perturbing the ambient plasma is also presented. The experimental results presented in this dissertation are the first laboratory measurement of dusty surface charging in a plasma. 38 Chapter 2: Experimental Setup and Surface Potential Diagnostics To determine the effects of dust on the charging of both conducting and insulating surfaces, it was necessary to create both a mesothermal plasma environment and a photoelectron environment. Additionally, it was crucial to establish a method to measure the surface potential of dielectric materials without perturbing the ambient plasma environment. This chapter details the vacuum facility, plasma and UV sources, plasma diagnostics, and the materials necessary to perform the experiments. The plasma parameters of the mesothermal plasma and photoelectron plasma environment are presented in their respective chapters. Details on the development of the embedded wire technique utilized to measure dust and insulating surface potentials are discussed in Chapter 3. 2.1 Vacuum Facility Figure 2.1: Primary vacuum chamber All experiments were conducted in a simulated space environment inside a cylindrical, stainless steel vacuum chamber. It measures 91.5 cm in diameter and 122 cm in length. An 2.2 Electron Bombardment Gridded Ion Source 39 Alcatel mechanical pump was used for roughing, and a CVI-TM500 cryogenic pump with a pumping speed of 8,500 L/s brought the chamber to high vacuum. The floor pressure is 1 10 7 Torr and maintains at 3 10 6 Torr with 2.5 sccm of argon gas flow. The chamber is equipped with one 8” Conflat port, six ISO-200 ports, and 11 KF-40 ports. 2.2 Electron Bombardment Gridded Ion Source (a) 4 cm ion source (b) Source electrical configuration Figure 2.2: Ion source and electrical configuration A 4 cm diameter electron bombardment gridded ion thruster with a hot filament neu- tralizer was used to create a mesothermal plasma. To generate plasma flow, argon gas flows through the back of the ionization chamber, and thermal electrons emitted from a hot tung- sten filament ionize the neutral argon gas. A magnetic field, created by ring magnets and a back magnet, confines the electrons along magnetic field lines to enhance collisions and ionization with the neutral gas molecules. The ionization chamber is biased to 1100 V above ground, and the anode cup is biased 50 V higher than the ionization chamber to absorb any low energy electrons that exist after collisions. This maintains continuous ionization. Ion optics accelerate the ions, generating a mesothermal plasma to simulate average solar wind conditions, and a hot-filament neutralizer is placed directly downstream of the source exit plane to reduce space charge effects and arcing. The source is contained in a grounded enclosure to screen the accelerated beam from the high voltage internal components and to 2.3 Ultraviolet Radiation Source 40 prevent potential perturbations. A photograph of the ion source inside the grounded enclo- sure is shown in Figure 2.2(a). The plastic tubing through which argon flows can be seen going into the back plate, and each of the tan wires is connected to an individual power source outside of the vacuum chamber to bias each component to its respective bias volt- ages. The black cylinder to the right of the photograph is the ionization chamber, where the plasma is generated. 2.3 Ultraviolet Radiation Source Figure 2.3: A single UV-C germicidal bulb Two commercial germicidal 25 W UV-C low-pressure bulbs with quartz sleeves, shown in Figure 2.3, were used to generate UV radiation. UV radiation is generated isotropically by exciting mercury sealed inside the low-pressure bulbs. At low pressure, mercury emits energy at 185 nm. Using Equation 2.1, where h is Planck’s constant, c is the speed of light, and is wavelength, this corresponds to photons with energies of 6.7 eV. E = hc (2.1) 2.4 Plasma Diagnostics 41 Figure 2.4: Traversing system 2.4 Plasma Diagnostics A probe suite had been developed to determine the plasma field parameters, such as ion and electron density, electron temperature, and plasma potential. 2.4.1 Data Acquisition and Traversing System A3-Dtraversingsystemwasusedtomountthediagnosticprobes. Thetraversingsystemuses stepper motors to move and has a linear resolution of 1 mil. Each motor can be controlled individually through LabVIEW, or a path file may be input to command the traversing system to move in a specific design. Data is collected at each point that the traversing system moves to. The data acquisition system was used to send and collect signals, which were sent back to the computer and written to a text file for post processing. 2.4.2 Langmuir Probe Utilizing the concept of surface charging, Langmuir 1961 [73] developed the concept of probe theory, and Bohm 1949 [74] and Allen, Boyd, and Reynolds 1956 [75] later refined the theory. To this day, it is still being studied extensively [76] [77] [78] [79] [80]. Probe theory is utilized 2.4 Plasma Diagnostics 42 to obtain the current-voltage (I-V) characteristics of a Langmuir probe immersed in plasma as the applied bias voltage of the probe is swept from a negative to a positive potential. The generated I-V curve can then be used to determine the electron temperature, T e , and electron density, n e . Figure 2.5: Langmuir probe and electrical configuration The characteristic length of the Langmuir probe used in this study was 6.35 mm and was greater than the Debye length of the ambient plasma environment, which was approximately 1 to 2 mm. Therefore, the thin sheath condition applied and could be used to calculate the electron density. The Langmuir probe was biased from -60 V to 60 V, and the voltage drop across a resistor was recorded for each bias voltage, as shown in Figure 2.5. This generated an I-V curve, as shown in Figure 2.6. T e and n e was determined utilizing Equations 2.2 and 2.3, respectively, where I e;sat is the electron saturation current and A probe is the probe collecting surface area. The plasma potential, p , could also be determined by the Langmuir probe, but its accuracy is low due to estimation errors. T e = 1 slope(T e line) (2.2) n e = I e;sat eA probe r 2m e kT e (2.3) 2.4 Plasma Diagnostics 43 Figure 2.6: Example of I-V curve generated by Langmuir probe data Figure 2.7: Faraday probe and electrical schematic 2.4.3 Faraday Probe AFaradayprobedeterminestheiondensitybymeasuringtheionfluxtoafixedareacollector. A stainless steel, nude Faraday probe, shown in Figure 2.7 was used for this experimental investigation. The collecting surface was set either perpendicularly or parallel to the plasma flow direction to collect normally flowing drifting ions and axially flowing thermal ions, and a concentric guard ring was placed around the collecting surface to create a uniform sheath and to minimize edge effects. The collecting surface outer diameter was 8.0 mm, and the guard ring outer diameter was 15.0 mm. The collecting surface and guard ring were both biased to -20 V to repel electrons and to ensure ions were the only species collected. The ion current density, J i , was determined by measuring the the voltage drop across a resistor 2.4 Plasma Diagnostics 44 and dividing it by the collecting surface area. Given the relations in Eq. 2.4 and 2.5, where 0 is the beam accelerating voltage, n i could be determined. J i =en i v i (2.4) 1 2 m i v 2 i =e( 0 p ) (2.5) n i = J i e r m i 2e( 0 p ) (2.6) 2.4.4 Emissive Probe Figure 2.8: Emissive probe and electrical schematic An emissive probe, shown in Figure 2.8, was used to measure the plasma potential, p . It consists of an exposed electrode and is electrically heated to the point of strong thermionic electron emission. Using the inflection-point method, the plasma potential can be determined by sweeping the emitting probe’s bias from large negative to large positive voltages and measuring the collected current at each bias voltage. Smith et al., 1979 [81] showed that the inflection point of the measured I-V curve is a good measure of the plasma potential. In the absence of a significant magnetic field or large density gradient, space charge effects are negligible and can be ignored. Sheath potentials can be measured utilizing this method. The emissive probe used in this experimental investigation had a probe radius of curvature of 2 mm with a tungsten filament diameter of 0.33 mm. 2.5 Lunar Regolith Simulant 45 2.4.5 Non-contacting Electrostatic Voltmeter Figure 2.9: Trek ESVM non-contacting surface potential measurement method A commercial Trek 323-L-CE non-contacting electrostatic voltmeter (ESVM) and model 6000B-8 side-view probe with a voltage range of 100 V, an accuracy of 50 mV, and a response time of 300 ms were used to measure surface potentials. This non-contacting electrostaticvoltmeter(ESVM)isavibratingcapacitiveprobethatdeterminessurfacepoten- tials with a current nulling method. A capacitor is created between the sample surface with potential s , and the probe head, with potential V. The probe head contains an electrode, which vibrates with frequency, !, and amplitude, d 1 . This forms a time-dependent gap, d =d 0 +d 1 sin(!t), where d 1 <d 0 between the sample surface and electrode, and when the electrode potential is equal to s , current stops flowing to the electrode. This determines the sample surface potential. The non-contacting probe ensures the surface is not disturbed when performing its measurement. Figure 2.9 illustrates the probe-surface capacitance and presents a side view of the probe above a sample surface. 2.5 Lunar Regolith Simulant Surface charging is highly dependent on material properties. To determine how dust affects spacecraftsurfacecharging,alunarregolithsimulant,JSC-1A,wasused. JSC-1Awascreated by the Johnson Space Center from glass-rich basaltic tuff mined from a commercial quarry at the Merriam Crater in Arizona. Its composition has been measured to be 46.2% SiO 2 , 17.1% Al 2 O 3 , 11.2% FeO, 9.43% CaO, 6.87% MgO, 3.33% Na 2 O, 1.85% TiO 2 , and other trace oxides [82]. JSC-1A possesses low electrical conductivity and dielectric loss, similarly 2.6 Gore-Tex 46 Figure 2.10: JSC-1A lunar regolith to lunar regolith, which provides confidence that experimental results obtained with JSC-1A will exhibit electrostatic trends similar to that of actual lunar regolith [83]. This can assist in developing charging models that can later by applied to the lunar surface. The JSC-1A dust grains used in all experiments were sifted to be between 60 and 100 m in diameter, and its relative permittivity is between 3.6 to 4.2. 2.6 Gore-Tex The outermost layer of the space suit is a thermal micrometeoroid garment shell. This material consists of a double woven fabric that is Gore-Tex on the outside and Nomex and Kevlarwoventothebacksidefortearresistance. Gore-Texisanon-flammablematerialmade of polytetrafluoroethylene (PTFE) and is used on the outer layer of space suits to protect the suit from any potential ignition sources, to withstand constant UV light exposure, and to break up micrometeroids and debris. Its relative permittivity is 2.0 [84]. To determine the effects of surface charging on a dusty insulating surface, a sample of this space suit material, developed by ILC Dover, was used. 47 Chapter 3: Development and Demonstration of the Dielectric Surface Potential Measurement Method Various contacting and non-contacting surface potential measurement techniques have been developed to measure the surface potential of dielectric surfaces immersed in plasma. How- ever, none of these methods have been capable of directly measuring the surface potential without perturbing the ambient plasma environment. In this chapter, a novel method to measure the surface potential of dielectric materials is presented. This method has the capa- bility to measure dust layer surface potentials as well as the surface potentials of insulating materials, and it is the foundation for subsequent dusty surface charging experiments. The effects of dust accumulation on the charging of a conducting plate are presented to demon- strate this new technique. 3.1 Surface Potential Measurement Techniques and Dif- ficulties There are multiple ways to measure the floating potential of a conducting surface in plasma. Some include measuring the voltage drop across a resistor or measuring the conducting sur- face’s current-voltage characteristics. There are also several methods of indirectly measuring the surface potential of dielectric surfaces in plasma. A popular method is to use an emissive probe to measure sheath potentials and to interpolate the surface potential. The other is to employ the use of a non-contacting electrostatic voltmeter (ESVM) to directly measure 3.1 Surface Potential Measurement Techniques and Difficulties 48 surface potentials inside a vacuum chamber. Both of these methods, however, introduce imprecision or difficulties. Sheath potential measurements made by emissive probes have been used to estimate the potential of surfaces [19] [85]. Though emissive probes are capable of measuring sheath potentials and surface potentials can be identified through interpolation, this method does not directly measure the potential of the surface. Additionally, if not carefully controlled, electrons emitted by emissive probes can alter the space charge condition within the sheath and can also alter sheath potentials. A second method to measure surface potentials is to use a non-contacting ESVM. This method is commonly used when determining ESD properties on solar arrays [70] [86] [87] [88] [89]. During solar array ESD testing, the solar arrays are charged by an electron beam gun to large potentials. The electron beam gun is then shut off inter- mittently to allow a non-contacting ESVM to scan the solar array surface and to measure surface potentials. The non-contacting ESVM must be shielded and cannot be used while the electron beam gun is in use; otherwise, the probe itself will build up charge and skew the measurements. Because the electron beam gun emits a single plasma species, the potential of the insulating surface after shutting off the electron beam gun is similar to what it was when the electron beam gun was on, and the measurements made by the non-contacting ESVM only need to be offset by the charge decay of the surface. When measuring the potential of a surface in a mesothermal plasma, however, the use of the non-contacting ESVM after the shut off of the plasma source and the neutralizer is not a valid method. The large difference between drifting ion and thermal electron velocities (electrons are9x faster than argon ions in the mesothermal plasma generated by the electrostatic gridded ion source in this study) creates transient effects when the plasma source and the neutralizer are shut off. If the plasma source and the neutralizer are turned off simultaneously, electrons are quickly absorbed, allowing the slower ions to dominate the plasma environment and charge surfaces more positively than their steady state surface potentials. If the shut off procedure is staggered and the neutralizer is turned off after the 3.2 Embedded-Wire Method 49 plasma source, electrons dominate the environment and charge surfaces negatively. There are no methods to counteract the transient effects from altering steady state potentials. If the plasma source must be on to preserve the surface potentials, it is impossible to accurately measure the surface potentials with the non-contacting ESVM in the vacuum chamber. However, the non-contacting ESVM is the most accurate method to directly mea- sure dielectric surface potentials without being in direct contact with the surface. Therefore, it was necessary to develop a method to directly measure surface potentials with the non- contacting ESVM placed outside the vacuum chamber while the plasma source was still running. 3.2 Embedded-Wire Method This method utilizes conducting wires embedded within the dielectric material and a Trek non-contacting ESVM. It is used in this dissertation to directly measure the potential of dust layers and insulating surfaces in plasma and the potentials within the dust layer itself. Previously, Green et al., 2006 [90] [91] measured the resistivity of dielectric samples and the electrostatic discharge of carbon composites by charging a dielectric surface, rotating the dielectric surface beneath a clean, conducting plate (witness plate) to form a capacitor, and measuring the induced potential on the witness plate. The witness plate was electrically connected to a conducting plate placed outside of the vacuum chamber, and a Trek non- contacting ESVM was used to measure the induced potential. In this method, similar to that of Green et al., 2006, one end of the embedded wire is connected to a measurement plate outside the vacuum chamber, which is measured by a Trek non-contacting ESVM, and the other end is placed at the dust layer surface or at different heights within the dust layer. Wires placed at the dust layer surface measure the surface’s floating potential, and those placed at various heights within the dust layer provide the qualitative potential profile inside the dust layer. By measuring the measurement plate outside the vacuum chamber with the plasma source running, there is no risk of altering steady state surface potentials. 3.2 Embedded-Wire Method 50 Figure 3.1: Sample plate measurement setup with Trek non-contacting ESVM probe and measurement plate placed outside of the vacuum chamber By using embedded wires as probes, the ambient plasma environment is not disturbed when the measurements are taken. To verify that the measurement plate potential measured by the Trek ESVM matched thatofanembeddedwireinsidethevacuumchamber, thechargingofacleanaluminumplate electrically connected to a wire in a mesothermal plasma was measured. The clean aluminum plate was electrically connected to the measurement plate outside the vacuum chamber with a PTFE insulated wire, as shown in Figure 3.1, immersed in the plasma generated by the ion source, and its floating potential was measured by the Trek ESVM. The PTFE insulated wire used to conect the sample plate and measurement plate is the same type of wire used for the embedded-wire method. Two additional floating potential measurement techniques were utilized to compare and validate the Trek ESVM measurement. The first technique was the floating point method. According to Ohm’s Law, the potential drop, V, across a resistor, R, is equal to the product of the current,I, through the resistor and the resistance (Eq. 3.1). By measuring the total current collected by the aluminum plate, the voltage drop across a 119.1 k resistor connected in series to the aluminum plate was calculated. Because one end of the resistor was connected to the aluminum plate and the other end was connected to ground, the potential drop across the resistor was equal to the floating potential of the aluminum plate. The second technique was to measure the I-V curve (Fig. 3.2) of the plate. According to current balance, the aluminum plate will float at a potential where the total ion and electron current collected is equal, resulting in zero current [78]. 3.2 Embedded-Wire Method 51 Figure 3.2: I-V trace of target aluminum plate V =IR (3.1) Table 3.1 shows the results of the three different floating point measurement techniques with respect to ground. The error on each floating potential measurement technique was calculated by dividing the standard deviation of the measured potentials by the square root of the sample size. Each measurement technique was run 12 times. The average floating potential measured by all three techniques was 1.94 V, and the standard deviation of the three techniques was 0.09 V. Given that each of the measurement techniques were within one standard deviation of the average floating potential of all three techniques, the results confirmed that the floating potential of a wire inside the vacuum chamber could be measured by electrically connecting the wire to a measurement plate outside of the vacuum chamber and measuring the measurement plate with a Trek ESVM. To ensure that the potential drop across the wire was negligible, the target plate was artificially biased by a power source, and the potential of the measurement plate was measured by the Trek ESVM. The potential drop across the wire was less than a hundredth of a volt. 3.3 Embedded-Wire Method Demonstration 52 Table 3.1: Floating potential measurements Method Floating Potential [V] Trek ESVM 2.030:03 I-V 1.920:03 Floating Point 1.870:02 Figure 3.3: Schematic showing the layout of the plasma source, sample plate, and the plasma flow field scan region 3.3 Embedded-Wire Method Demonstration To validate the embedded-wire method, the charging of an aluminum plate covered by a thin layer of JSC-1A was measured. The target sample was a 15.2 cm 10.2 cm aluminum plate placed 17.8 cm downstream of the plasma source (x-direction) and 2.54 cm below the centerline of the plasma source. The plate orientation was parallel to the plasma beam direction (see Fig. 3.3). Because the cold ion plasma beam had about a 0 angle of attack with respect to the target sample, the plasma current collected at the plate surface was approximately that of cold ions and thermal electrons. 3.3.1 Plasma Environment The plasma environment was measured in a region 12.7 cm in the axial direction by 7.62 cm in the vertical direction, 2.54 cm above the sample plate (Fig. 3.3) with an electrostatic Langmuir probe and Faraday probe. The scanning area was divided into 91 measurement points with a spatial resolution of 1.27 cm by 1.27 cm. An emissive probe scanned a region 3.3 Embedded-Wire Method Demonstration 53 Figure 3.4: Photo of experimental setup inside the vacuum chamber 12.7 cm in the axial direction by 7.86 cm in the vertical direction, 0.74 cm above the sample plate to capture the ambient plasma potential and sheath potential above the sample. The scanning area was divided into 286 measurement points with a spatial resolution of 0.06 cm by 1.27 cm within the sheath and 1.27 cm by 1.27 cm above the sheath. Figure 3.5 shows the measured contour plots of the plasma potential, p , current density, J i , ion density, n i , electron density, n e , space charge, n i n e , and electron temperature, T e . Note that the closest measurements to the target sample were made at 25.4 mm above the surface, which aligned with the centerline of the plasma beam. The measured potential at the source center line above the leading edge of the plate (x = 179 mm, z = 25.4 mm) was used as the reference ambient potential. Perturbations in the measured plasma parameters were due to the non-steady plasma source. Throughout experimentation, the beam current and neutralizer current must constantly be readjusted to maintain the average plasma beam current at 10 mA and the neutralizer current at 11 mA. It should be noted that the plasma environment is measured and presented for every experimentconductedinthisdissertation. BecausetheDebyelengthoftheplasmaisapprox- imately 2 mm, small differences when aligning the plasma source with the target sample between experiments can create differences in the plasma environment 178 mm downstream of the plasma source exit. For purposes of this dissertation, however, knowing what the 3.3 Embedded-Wire Method Demonstration 54 (a) p , plasma potential (b) J i , ion current density (c) n i , ion density (d) n e , electron density (e) n i n e , space charge (f) T e , electron temperature Figure 3.5: Plasma environment above the sample plate. The centerline of the plasma beam is at z = 25.4 mm. The reference potential is the potential at x = 179 mm, z = 25.4 mm. The potential values plotted are offset by the reference potential Table 3.2: Average ambient plasma parameters at various x-positions along beam centerline x-position [mm] p [V] n i [m 3 ] n e [m 3 ] T e [eV] 177.8 0.0 9.110 13 5.410 13 2.4 228.6 -2.4 7.310 13 3.810 13 1.7 279.4 -3.9 6.410 13 3.110 13 3.2 330.2 -4.7 6.110 13 2.410 13 2.1 plasma environment is for each experimental setup is more important than maintaining the exactplasmaenvironmentforeachexperiment; thisisbecauseanalyticalresultsderivedfrom the measured ambient plasma characteristics are used to understand the charging process and to compare with experimental results. The average ambient plasma characteristics for all of the experimental setups will be discussed and compared in Chapter 7. 3.3 Embedded-Wire Method Demonstration 55 Table 3.2 shows the average measured plasma potential, the ion density, the electron density, and the electron temperature at various x-positions along the beam centerline. From these measurements, the Debye length in the plasma environment was approximately D = 2 mm. For the setup considered here, the standard 1-D space-charge limited current collection analysis can be applied to estimate the surface floating potential with respect to the ambient, s , and thesheath thickness,d sh . Alongthe z-direction, a pre-sheathaccelerates theambient cold ions to satisfy the Bohm sheath condition. Hence, the ions enter into the sheath with a Mach number, M = v iz =C s = 1, where C s = p kT e =m i is the ion acoustic velocity. Therefore, the ion current collection along the z-direction towards the plate is approximately J i 'en io C s . The electron current outside the sheath is the thermal current. Hence, the 1-D current balance condition at the surface is en io C s =en eo r kT e 2m e exp( e s kT e ) (3.2) and the sheath thickness is given by d sh D ' ( 4 p 2 9 ( e s kT e ) 3=2 ) 1=2 (3.3) Using the average ion density, electron density, and electron temperature from Table 3.2, the surface potential should be s '9 V with respect to the ambient reference with a sheath thickness of d sh ' 4 mm. 3.3.2 Dust-Covered Conducting Plate The sample plate was an aluminum plate that was electrically floating with respect to the plasma environment. In this study, the floating potential of a clean plate was compared to the floating potential of a plate that was covered by a layer of dust. The potential at the dust layer surface was obtained using the embedded-wire method. Five 22-gauge PTFE high-temperature stranded wires were inserted through the bottom 3.3 Embedded-Wire Method Demonstration 56 Figure 3.6: Schematic of diagnostic wire positions on the sample plate Figure 3.7: Photo of a sample plate covered by a layer of JSC-1A regolith simulant of the plate and along the centerline of the aluminum plate, as shown in Figure 3.6. The wire insulation electrically isolated the wires from the aluminum plate, and the tips of each wire were exposed to the plasma. Because the floating potential at the dust layer surface is determined by the current balance condition at the surface, the wires with exposed tips at the same level as the dust surface floated to the same potential as the dust layer surface. Additionally, embedded wires were placed inside the dust layer with tips exposed directly to the dust. These embedded wires provided qualitative information on the potential inside the dust layer but not the actual dust potentials as the contact resistance between the wire and the dust was not known. Similarly to the aluminum plate, each wire was individually connected to a respective aluminum measurement plate placed outside the vacuum chamber, and the non-contacting ESVM measured the potential of each wire. 3.3 Embedded-Wire Method Demonstration 57 Figure 3.7 presents a photo of an aluminum plate completely covered by a layer of JSC- 1A lunar regolith, and Figure 3.8 illustrates the wire tip position above the plate (along the z-direction) as well as the two different dust layer thicknesses considered in the experiments. Four different wire heights (listed in Table 3.3) were used to measure the potential both inside the dust layer and just above the dust layer surface. Dusty plate 1 had a dust layer thickness of 0.16 cm, which corresponds to the h2 wire level, and dusty plate 2 had a dust layer thickness of 0.32 cm, which corresponds to the h4 wire level. In each test, all of the wires were placed at the same height. Hence, four different sample plates were used for each dusty surface. To ensure all wire tips were placed at the same height on each sample plate, four individual markers with thicknesses of h1, h2, h3, and h4, respectively, were made. When inserting each wire through the bottom of the aluminum plate, the marker was used to line the tip of the wire with the top of the marker. Using the same marker for all wires on each sample plate ensured uniformity across all of the wires. To guarantee that the dust thickness was uniform across all four sample plates, two stencils, 0.16 cm and 0.32 cm thick, respectively, were developed. The stencil was placed around the aluminum plate, dust was set onto the aluminum plate’s surface, and a straight-edge was used to run along the stencil’s upper surface, thereby removing any excess dust and ensuring that the dust thickness on the aluminum plate was uniform. The accuracy of the dust layer thickness with respect to its respective height was based on the stencil thickness tolerance, which was0.005 cm. Before each case was tested, the dust samples were held under vacuum and baked with a Watlow polyimide sheet heater to outgas residual moisture. Table 3.3: Wire tip height above the sample plate Wire Level Height h1 0.08 cm h2 0.16 cm h3 0.24 cm h4 0.32 cm To ensure that the charging of the Teflon wire insulation and the Watlow polyimide sheet heater did not affect the charging of the samples, the floating potentials of a clean aluminum plate, a clean aluminum plate with a Watlow polyimide sheet heater, a clean aluminum plate 3.3 Embedded-Wire Method Demonstration 58 (a) Wire height setup (b) Dusty plate 1 (c) Dusty plate 2 Figure 3.8: Illustration of diagnostic wire height in experimental setup and dust layer thick- ness considered in experiment withembeddedPTFEwires, andacleanaluminumplatewithbothaWatlowpolyimidesheet heater and embedded PTFE wires were measured. It was found that the floating potential of all four samples were identical, which confirmed that charge stored on the dielectrics did not distort the measurements. In addition, wires embedded in the dust layer were not exposed to the plasma and would not affect the charging of the samples. The potential of two dust layer thicknesses were measured. The first layer had a thickness of 0.16 cm. In this setup, the tip of the h1 wires were embedded in the dust layer, the tip of the h2 wires were at the dust surface, and the tips of the h3 and h4 wires were above the dust surface. Then, another layer of JSC-1A was applied so that the total dust layer thickness was 0.32 cm. In this setup, the tips of the h1, h2, and h3 wires were all embedded within the dust layer, and the tips of the h4 wires were at the dust surface. 3.3 Embedded-Wire Method Demonstration 59 Figure3.9: 1-Dpotentialprofileoftheambientplasmaandsampleplatealongthez-direction at x = 254.0 mm 3.3.3 Dusty Surface Charging Figure 3.9 presents the 1-D potential profile of the ambient plasma and the sample plate along the z-direction at x = 254.0 mm. The potentials above the dust layer were measured by the emissive probe, and the poten- tials inside or near the dust layer were measured by the embedded wires. The floating potential of the aluminum plate potentials was measured by the non-contacting electrostatic ESVM. Figure 3.10 shows the floating potential of the target plate along the x-direction. For comparison, the potentials of the clean plate and the two dusty plates considered are also plotted. The potentials measured for the plate and that of the dust surface represent the actual floating potential, while the potentials measured for locations inside the dust layer are considered to be qualitative. Nevertheless, the potential profile shows that the potential inside the dust layer becomes more negative with increasing depth. The floating potential of the conducting plate and the average floating potential of the dust surface for all three cases are shown in Table 3.4. Not surprisingly, the potential of the clean plate and of the two 3.3 Embedded-Wire Method Demonstration 60 (a) Clean plate (b) Plate with h2 dust thickness layer (c) Plate with h4 dust thickness layer Figure 3.10: Sample potential within dust layer measured by embedded wires 3.3 Embedded-Wire Method Demonstration 61 Table 3.4: Aluminum plate floating potential and dust surface potential Cases Dust Dust Surface Plate Floating % Plate Potential Thickness Potential [V] Potential [V] Change vs. Clean Plate Clean Plate - - -12.740:02 - h2 Dust Layer 0.16 cm -12.220:12 -13.890:02 -9.4% h4 Dust Layer 0.32 cm -13.380:14 -15.130:08 -18.9% dusty surfaces were all very similar. This was to be expected because the surface charging potential is controlled by the current balance condition. The error bars on the dust layer potentials were calculated by dividing the standard deviation by the square root of the sample size. Each test setup was run four times, and each wire was measured 18 times per run for a total of 72 measurements per wire per test setup. 3.3.4 Dust Layer Capacitance The results show that the existence of a dust layer drives the plate potential more negatively with respect to the ambient potential. This is to be expected. Assuming the dust layer is a parallel plate capacitor, the relation between the dust surface potential and the plate potential can be modeled by C parallel = 0 rd A d (3.4) Q =C( s plate ) (3.5) whereA is the overlapping area between the dust layer and aluminum plate,d is the dust thickness, and plate is the plate potential. Because s is determined by current collection from the ambient plasma, the charge stored in the dust layer drives plate more negatively. Notsurprisingly, asthedustlayerthicknessincreased, thedifferencebetweenthedustsurface potential and the plate potential also increased linearly, as was shown in Figure 3.9 and in Table 3.4. 3.4 Summary and Conclusion 62 Table 3.5: Total charge on dust layer and charge per dust grain Dust Thickness Q total [e] [e=m 2 ] Q d [e] h2 Dust Layer 3:3 10 9 2:2 10 11 1:7 10 3 h4 Dust Layer 2:4 10 9 1:6 10 11 1:3 10 3 From the measured potential difference between the dust layer surface and plate, the charge deposited on the dust layer, Q, can be estimated from Eq. 3.5. Because there is little charge transfer between the dust grains [92], the collected charge is distributed on the dust layer surface. Therefore, the average charge accumulated on a single dust grain with a radius r d on the dust surface is Q d =r 2 d ; = Q A = C( s plate ) A (3.6) Table 3.5 shows the estimated total charge deposited on the dust layer,Q total , the average charge density of the dust surface,, and the average charge per dust grain on the dust layer surface, Q d . If the dust grains were isolated, the charge accumulated on a single, isolated dust grain with the same radius and dust potential would be Q d;iso = 4 0 rd r d s when assuming a spherical dust grain. This calculation would lead to a charge per dust grain of approximately Q d;iso 1:510 6 e, where e is defined as the elementary charge (1.610 19 C). This is about three orders of magnitude larger than that of a charge per dust grain on a dusty surface, which is in agreement with the results in [93], where they measured dusty layer charging in a localized N + plasma wake and found that the charge stored by an individual dust grain on a regolith surface was approximately two-orders of magnitude less than that of a single, isolated dust grain at the same dust charging potential. 3.4 Summary and Conclusion A laboratory study was carried out to verify the embedded wire method and to measure the charging of a conducting surface covered by a thin layer of lunar dust simulant, JSC-1A, in plasma. The floating potentials of the dust layer and the plate beneath the dust layer 3.4 Summary and Conclusion 63 were obtained by utilizing embedded wires connected to a measurement plate outside the vacuum chamber and measured by a Trek non-contacting ESVM. The results showed that the floating potential of a dust-covered conducting plate is dependent upon both the plasma current collection and the dust layer thickness. The current balance condition controls the dust surface potential, and the dust layer acts as a capacitor between the exposed dust surface and the plate underneath the dust. Hence, the dust layer will drive the potential of the plate beneath it further negatively with respect to the ambient. This is the first known direct method of measuring the charging of a dust-covered surface in a plasma. The embedded-wire method is capable of measuring dielectric surface potentials in a plasma without perturbing the ambient field or the surface itself and can be configured in multiple ways. For the work presented in this dissertation, only embedded wires were used to measure surface potentials, but future work may consider using other types of probes, as will be discussed in recommended future work in Chapter 8. 64 Chapter 4: Plasma Charging of Dusty Conducting Surface: Dust Coverage Effects on Spacecraft Charging This chapter presents a study on how both the coverage pattern and the quantity of the dust affect the charging of a conducting surface in a mesothermal plasma. Chapter 3 considered a surface covered by a uniformly distributed this layer. The focus of this chapter is on dust coverage pattern. The floating potential of the dust layer surface and of the conducting surface beneath it are measured directly utilizing the embedded-wire method, and a series of 3-D,fullykineticparticle-in-cell(PIC)simulationsisdevelopedforcomparisonandvalidation of the experimental results. 4.1 Experimental Setup The experimental method discussed in Chapter 3 is applied to determine the effects of dust coverage on the charging of an aluminum plate in a mesothermal plasma of cold ions and thermal electrons at 0 angle of attack. This setup simulates a plate placed at the lunar terminator surface under the average solar wind condition plasma. The target sample was an electrically-floating 15.2 cm 10.2 cm aluminum plate covered with JSC-1A dust grains. It was placed 17.8 cm downstream of the plasma source (x-direction) and 2.54 cm below the centerline of the plasma source. The plate orientation was parallel to the plasma beam direction (see Fig. 4.1). 4.2 Plasma Environment 65 Figure 4.1: A schematic showing the layout of the plasma source, sample plate, and the plasma flow field scan region 4.2 Plasma Environment The plasma environment was measured in a region 15.24 cm in the axial direction by 6.35 cm in the vertical direction, 1.91 cm above the sample plate (Fig. 4.1) with an electrostatic Langmuir probe and a Faraday probe. The scanning area was divided into 78 measurement points with a spatial resolution of 1.27 cm by 1.27 cm. An emissive probe scanned a region 15.24 cm in the axial direction by 8.15 cm in the radial direction, 0.13 cm above the sample plate to capture the ambient plasma potential and sheath potential above the sample. The scanning area was divided into 260 measurement points with a spatial resolution of 0.06 cm by 1.27 cm within the sheath and 1.27 cm by 1.27 cm above the sheath. Figure 4.2 shows the measured contour plots of the plasma potential, p , current density, J i , ion density, n i , electron density, n e , space charge, n i n e , and electron temperature, T e . Note that the closest measurements to the target sample were made at 19.1 mm above the surface, and the centerline of the plasma beam was 25.4 mm above the surface. The measured potential at the source center line above the leading edge of the plate (x = 179 mm, z = 25.4 mm) was used as the reference ambient potential. Table 4.1 shows the average measured plasma potential, ion density, electron density, and electron temperature at various x-positions along the beam centerline. The Debye length in the plasma environment was approximately D = 2 mm. 4.2 Plasma Environment 66 (a) p , plasma potential (b) J i , ion current density (c) n i , ion density (d) n e , electron density (e) n i n e , space charge (f) Te, electron temperature Figure 4.2: Plasma environment above the sample plate. The centerline of the plasma beam is at z = 25.4 mm. The reference potential is the potential at x = 179 mm, z = 25.4 mm. The potential values plotted are offset by the reference potential Table 4.1: Average ambient plasma parameters at various x-positions along beam centerline x-position [mm] p [V] n i [m 3 ] n e [m 3 ] T e [eV] 177.8 0.0 8.610 13 3.010 13 1.3 228.6 -1.3 5.410 13 1.510 13 2.6 279.4 -2.4 5.310 13 1.010 13 3.4 330.2 -3.5 4.310 13 9.310 12 2.7 Using the average ion density, electron density, and electron temperature from Table 4.1 and applying the standard 1-D space-charge limited current collection analysis for cold ions and thermal electrons to estimate the floating potential of a clean conducting plate, s , (Eq. 3.2) and the sheath thickness (Eq. 3.3), the surface potential is approximately s ' 10:6 V with respect to the ambient reference, and the sheath thickness is d sh ' 4 mm above the sample plate. 4.3 Dusty-Plate Samples 67 (a) Clean plate (b) 1 sq (c) 7 sq checkerboard (d) 7 sq 3 0 1 0 3 (e) 7 sq 3 3 1 0 0 (f) 7 sq 0 0 1 3 3 (g) 14 sq (h) Full dust coverage Figure 4.3: Dusty plate sample setups 4.3 Dusty-Plate Samples To examine the effects of dust coverage on surface charging, seven different dusty-plate samples were created with JSC-1A (Fig. 4.3), and their charging results were compared to the clean plate condition. To create each dusty-plate sample, a grid was placed onto the aluminum plate, and seven different dust layer patterns were laid onto the conducting plate. The effect of dust positioning was examined by filling the same number of grid squares but changing which dust squares were filled. The effect of dust coverage was examined by varying the number of grid squares filled by dust. A 0.16 cm thick dust layer was used to fill each grid square. The aluminum plate’s floating potential and the surface potential of the dust were obtained utilizing the embedded-wire method. Each dust sample configuration will be referred to by the position and number of grids squares filled by dust, as indicated in the labels in Figure 4.3, and grid squares filled with dust will be referred to as dust squares. 4.4 Current Collection-Capacitor Model Priortomeasuringthefloatingpotentialsofeachdustysample, acurrentcollection-capacitor model was created by combining the current-balance condition with the dust layer capacitor effects found in Chapter 3. This model was used to estimate the effects of the dust coverage 4.4 Current Collection-Capacitor Model 68 on the surface charging. The floating potential of the conducting plate due to current collection, cc , was first calculated using the 1-D space-charge limited current collection analysis for cold ions and thermal electrons on plate surfaces exposed to plasma (Eq. 3.2). Dust-square surface potentials, dust , were also calculated using a 1-D space-charge limited current collection analysis. A series of emissive probe measurements were taken at a distance of d EP = 0.13 cm above a full dust coverage surface in the X-Y plane to measure the dust layer sheath potentials, sheath . They were combined with the calculated floating potentials of each dust square to determine the local electric field,E field (Eq. 4.1). Gauss’ law (Eq. 4.2) was then used to calculate the charge stored on each dust square. Usingthemodelforaparallelplatecapacitor(Eq.3.4), eachdustsquarewasmodeledasa capacitor-in-paralleltodeterminethetotalchargestoredonthedustlayer, whichdetermined how negatively the total dust layer capacitance drove the conducting plate potential, V. This was combined with the calculated plate potential from current collection, cc , in Eq. 4.3 to determine the estimated conducting plate floating potential, s;est . E field = sheath dust d EP (4.1) Q A dust =E field 0 rd (4.2) s;est = cc + V (4.3) The floating potentials calculated by the current collection-capacitor model (Table 4.2) followed an expected trend; as dust coverage increased, the dust layer capacitance drove the potential of the conducting plate beneath it more negative, and the floating potential of a conducting plate completely covered by dust was the most negative. 4.5 Dusty-Surface Charging 69 Table 4.2: Dust coverage percentage and floating potential with respect to ambient plasma Dust Current Collection - Measured Simulation Coverage % Capacitor Model [V] Potential [V] Potential [V] No Dust 0 -10.61:1 -10.790:07 -10.610:19 1 sq 6.7 -10.71:0 -10.930:10 -11.100:24 7 sq checkerboard 46.7 -10.81:1 -12.000:15 -11.680:24 7 sq 3 0 1 0 3 46.7 -10.91:1 -11.810:06 -11.680:24 7 sq 3 3 1 0 0 46.7 -10.81:1 -10.970:06 -10.770:20 7 sq 0 0 1 3 3 46.7 -11.11:1 -10.660:07 -10.850:23 14 sq 93.3 -11.31:1 -10.870:13 -10.980:21 Full Dust Coverage 100 -11.41:1 -11.700:10 -11.370:20 4.5 Dusty-Surface Charging Table 4.2 presents the experimentally measured floating potentials of the clean plate and the seven different dusty-plate samples. The error bars on the measured floating potentials were calculated by dividing the standard deviation of the measured potentials by the square root of the sample size. Each test setup was run four times, and each sample plate was measured 12 times per run for a total of 48 measurements per dusty sample. The measured floating potential of the clean aluminum plate was -10.8 V, which is in good agreement with the analytically calculated floating potential of -10.6 V. It is clear that the accumulation of dust on a clean conducting plate drives the potential more negatively. However, it was surprising to see that the 1 sq. plate charged more negatively than the 7 sq. 0-0-1-3-3 sample plate, and the 7 sq. checkerboard and the 7 sq. 3-0-1-0-3 sample plates both charged more negatively than did the full dust coverage plate. Because the capacitance of the dust layers could only drive the potential of the conducting plate more negative by an amount proportional to its charge and thickness, these measurements made it clear that the dust pattern was affecting the current collection at the plate surface. 4.6 Simulation Study Approach 70 4.6 Simulation Study Approach 4.6.1 Simulation Model To help explain this phenomena observed in the experiment, a series of particle-in-cell (PIC) simulations was carried out. The simulation code used was the Immersed-Finite-Element PIC model developed at USC, the USC-IFEPIC [94][95]. The USC-IFEPIC is a 3-D fully- kinetic PIC model with capabilities to resolve the charging of complex dielectric objects in plasma. In the PIC method, ions and electrons were modeled as macro-particles and distributed “freely” in the entire simulation domain, and the code solved the electrostatic Poisson’s equation self-consistently. At each time step, the charges carried by simulation particles were deposited onto mesh nodes, Poisson’s equation was solved for the electric potential and for the field, the electric field was interpolated from the mesh nodes to update each particle’s velocity, and the equation of motion was integrated to update simulation particle positions. This cycle was iterated for a determined number of time steps until a self-consistent solution of both plasma and field was obtained. 4.6.1.1 Governing Equations The properties of electromagnetic fields are described by Maxwell’s equations, and for elec- trostatic interactions, they follow the form: rB = 0 (4.4) rB = 0 (4.5) rE = 0 (4.6) rE = 0 (4.7) 4.6 Simulation Study Approach 71 Given thatrE = 0, the electric potential, , can be modeled as E =r (4.8) and the electric field can be solved by Poisson’s equation: rE =rr =r 2 = 0 (4.9) For a collisionless, unmagnetized plasma under an electrostatic field, Newton’s second law and the Lorentz force determine the trajectory of each charged particle: F =qE =m dv dt (4.10) where F is force, q is electric charge carried by each particle, and v is velocity. The electric field, E, is calculated from electric potential, , by Eq. 4.8. For purposes of this research, the external magnetic field is neglected, which is valid for local plasma-surface interactions on the Moon [94]. 4.6.1.2 Normalization To reduce round-off errors when performing calculations with extremely large or extremely small numerical values, the governing equations are normalized by the reference parameters shown in Table 4.3. The dimensional quantities are normalized as ^ X = X X ref (4.11) Table 4.3: Normalization reference parameters L ref q ref m ref T ref ref n ref v ref t ref D e m e T e kT e =e n e v te 1=w pe 4.6 Simulation Study Approach 72 Figure 4.4: PIC loop performed at each timestep 4.6.1.3 PIC Loop To determine the plasma dynamics and field quantities self-consistently, the PIC simulation performs the following major steps, shown in Figure 4.4, at each timestep [23][95]: The charges carried by all of the simulation particles are distributed proportionally onto the mesh nodes depending on each particle’s position with respect to each node. The position of each particle with respect to the surrounding cell nodes determines the size of each “area,” and thus, the percentage of charge deposted onto each node, forming the right hand side of Poisson’s equation. Figure 4.5 illustrates this charge distribution. The red dot symbolizes a particle. Given that the particle is closest to the “Area 4” node, the “Area 4” node should receive the largest proportion of charge, which corresponds with Area 4. Similarly, the “Area 1” node is furthest from the particle, and the charge distributed onto its node is equal to Area 1. Poisson’s equation is solved for electric potential, , and field, E. 4.6 Simulation Study Approach 73 Figure 4.5: Charge distribution and deposition onto mesh nodes [23] The electric field at each particles’ position is interpolated from the mesh nodes and used to update the particles’ velocities The equation of motion is integrated to update the position of each particle 4.6.1.4 Surface Charging Objects in the simulation are modeled as an interface inside the computation domain via the immersed finite element (IFE) scheme. At each PIC step, charge deposition at the interface is determined by retrieving the “old” and “new” positions of each simulation particle and determining if the line between the two positions intersects the interface. If the particle path intersects the interface segment, the particle is collected, and the interface accumulates surface charge. At the end of each PIC step, the total accumulated surface charge in each interfaceelementisdividedbyitscorrespondingareatodeterminethesurfacechargedensity. The surface charge density is then used to update the charge on each node in the object interface, and the cycle is repeated. 4.6 Simulation Study Approach 74 Figure 4.6: Simulation domain setup 4.6.2 Simulation Domain Setup A domain size of 120 cells by 2 cells by 60 cells with a mesh size of 1.0 in each direction was used for a normalized domain size of 120 2 60, and a total of 2.410 6 particles were used. The plate interface was set at z = 7.9, and the dust surface was set at z = 9.9 (Fig. 4.6). Particles were pre-loaded into the simulation domain with a full Maxwellian distribution and drifted along the positive x-direction. Because the simulation must consider particles that flow into the domain due to thermal motion, particles were injected at X min , X max , and Z max at the boundaries. Particles were also absorbed at X min , X max , Z min , and Z max within each PIC iteration step. Reflection particle boundary conditions were applied at Y min and Y max . The plasma species (cold, drifting ions and thermal electrons) flowed through the simulation domain for a number of simulation steps with a time step size of 0.1 The normalized velocity parameters chosen for each plasma species are shown in Table 4.4. The normalized drifting, ^ V, and the normalized thermal, ^ v, velocities were normalized by the electron thermal velocity, and the simulation potentials were normalized by the average electron temperature measured in Figure 4.2(f). 4.7 Simulation Study: Dust Pattern Effects 75 Table 4.4: Normalized ion and electron velocities and temperature Species ^ n ^ V ^ v ^ T Electrons 1.0 0.109500 1.000000 1.000000 Ions 1.0 0.109500 0.000473 0.015000 4.7 Simulation Study: Dust Pattern Effects Each dusty plate was modeled by the simulation. Figures 4.7-4.9 present the steady-state simulation potential field, ion density, and electron density, respectively, for each dusty plate sample. The ion density results (Fig. 4.8) are particularly interesting. Results show that the trailing edge of each dust surface creates a plasma wake region due to expansion of the collisionless, mesothermal plasma flow. The steady state structure of this expansion is analogous to the Prandtl-Meyer expansion fan generated by a supersonic gas flowing around a corner [96], and the electric potential and ion density decrease as the plasma expands into the wake region. The presence of these “micro-wake” regions at the trailing edge of dust surfaces creates an electron dominant region, which increases the electron current collection of the conducting plate and drives the potential more negatively than that calculated by the current collection- capacitor model. These simulation results show that conducting surfaces with “patches" of dust accumulation, such as the 7 sq. checkerboard dusty-sample plate, can charge more negatively than those with more uniform dust layers, such as the 7 sq. 3-3-1-0-0 and 7 sq. 0-0-1-3-3 dusty-sample plates. Given the same dust thickness, “patchy" dust coverage can even charge conducting surfaces more negatively than conducting surfaces with more overall dust coverage, such as the 14 sq. and full-dust coverage samples. These results show that the plasma wake created at the trailing edge of 0.16 cm thick dust layers can have a greater effect than the dust layer capacitor on driving the potential of a conducting plate more negatively. This agrees well with the experimentally measured dusty sample floating potentials. 4.7 Simulation Study: Dust Pattern Effects 76 (a) Clean plate (b) 1 sq (c) 7 sq checker (d) 7 sq 3 0 1 0 3 (e) 7 sq 3 3 1 0 0 (f) 7 sq 0 0 1 3 3 (g) 14 sq (h) Full dust coverage Figure 4.7: Simulation dusty-sample potential field 4.7 Simulation Study: Dust Pattern Effects 77 (a) Clean plate (b) 1 sq (c) 7 sq checker (d) 7 sq 3 0 1 0 3 (e) 7 sq 3 3 1 0 0 (f) 7 sq 0 0 1 3 3 (g) 14 sq (h) Full dust coverage Figure 4.8: Simulation dusty-sample ion density 4.7 Simulation Study: Dust Pattern Effects 78 (a) Clean plate (b) 1 sq (c) 7 sq checker (d) 7 sq 3 0 1 0 3 (e) 7 sq 3 3 1 0 0 (f) 7 sq 0 0 1 3 3 (g) 14 sq (h) Full dust coverage Figure 4.9: Simulation dusty-sample electron density 4.8 Simulation Study: Dust Coverage vs. Dust Thickness 79 (a) Potential (b) Ion density (c) Electron density Figure 4.10: Thick checkerboard simulation (a) Potential (b) Ion density (c) Electron density Figure 4.11: Thick full dust coverage simulation 4.8 Simulation Study: Dust Coverage vs. Dust Thick- ness To confirm that the type of dust coverage on a conducting plate can have a greater effect on the charging of a dust-covered conducting plate than the thickness of the dust layer, a “thick" checkerboard and a “thick" full dust simulation were created. A domain size of 120 cells by 2 cells by 70 cells with a mesh size of 1.0 in each direction was used for a normalized domain size of 120 2 70, and a total of 2.410 6 particles were used. The plate interface was set at z = 7.9, and the dust surface was set at 17.9. The particle loading, boundary conditions, and normalized plasma parameters used in Chapter 4.6 were used in these simulations as well. 4.9 Summary and Conclusion 80 Table 4.5: “Thick" simulation results Plate Potential [V] Checkerboard -15.090:20 Full dust coverage -11.490:21 Figures 4.10 and 4.11 compare the “thick" checkerboard and “thick" full dust coverage cases, respectively. A more pronounced plasma wake region is created behind the trailing edgeofeachdustlayerinthe“thick"checkerboardsetupcomparedtothatofthethinnerdust layer simulation in Chapter 4.6. This drives the potential of the surface in the wake region even more negatively. Table 4.5 presents the floating potentials of the “thick" checkerboard and “thick” full dust plates and confirms that the type of dust coverage can have a greater effect on charging than dust layer thickness. 4.9 Summary and Conclusion A laboratory study was carried out to measure the charging of a conducting surface partially covered by varying quantities of lunar dust simulant, JSC-1A, in a collisionless, mesothermal plasma. The floating potentials of the dust layer and the plate beneath the dust layer were obtained utilizing the embedded-wire method. Results showed that patches of dust accumulation on a conducting plate can drive the potential of a dust-covered conducting plate more negatively than that of a conducting plate covered by a uniform layer of dust of the same thickness. This is due to the plasma wake regions generated at the trailing edge of dust patches, which contains a larger density of electrons than ions and increases electron current collection. Given uniform dust coverage, effects from plasma wakes are minimized and effects from dust layer capacitance dominate. Hence, the way in which the dust accumulates on a conducting surface has a more significant effect on surface charging than does the quantity of dust that accumulated. 81 Chapter 5: Plasma Charging of Dusty Insulating Surface: Dust Coverage Effects on Space Suit Charging Aside from dust accumulation on spacecraft surfaces, dust will also accumulate on space suits and insulating material surfaces. An unforeseen problem during the Apollo mission was dust contamination on sensitive equipment and on the space suits. The astronauts doc- umented the adhesion of dust to their space suits, which would be carried into the cabin of the lunar module, where the dust would make breathing difficult. The dust would also increase the wear and tear of the suits, and it would obscure vision as well [10]. Because insulating materials cannot distribute charge evenly over their surfaces, the surface charging of an insulating material, such as that found on space suits, gives rise to differential charg- ing. Clean, insulating surfaces exposed to the solar wind will charge to potentials following standard space-charge limited current collection analysis from ambient plasma (Fig. 5.1(a)). However, when insulating materials are present in a dusty environment, such as that found onthesurfaceoftheMoon, dustaccumulationwillaffecthowtheinsulatingmaterialcharges, and the dust can potentially increase the severity of the differential charging (Fig. 5.1(b)). In turn, this may increase the probability for arcing and discharge. 5.1 Experimental Setup This experimental method is applied to determine the effects of the JSC-1A lunar dust simulant coverage on the charging of Gore-Tex in a mesothermal plasma of cold ions and thermal electrons at 0 angle of attack. The target sample was a 7.6 cm 10.2 cm piece of Gore-Tex resting on a 15.2 cm 10.2 cm aluminum plate. The aluminum plate was 5.1 Experimental Setup 82 (a) Clean insulating surface (b) Dusty insulating surface Figure 5.1: Clean and dusty insulating surface flux l Figure 5.2: Schematic showing the layout of the plasma source, sample plate, and the plasma flow field scan region used as a base to mount the Gore-Tex sample. Given that the Debye length of the plasma environment was approximately 2 mm, the charging of the aluminum plate had negligible effect on the charging of the Gore-Tex and the dusty surfaces. The target sample was placed 17.8 cm downstream of the plasma source (x-direction) and 2.54 cm below the centerline of the plasma source. The plate orientation was parallel to the plasma beam direction (see Fig. 5.2). Because the cold ion plasma beam had about a 0 angle of attack with respect to the target sample, the plasma current collected at the plate surface was approximately that of cold ions and thermal electrons. 5.2 Plasma Environment 83 Table 5.1: Average ambient plasma parameters at various x-positions along beam centerline x-position [mm] p [V] n i [m 3 ] n e [m 3 ] T e [eV] 177.8 0.0 1.610 14 5.410 13 1.8 228.6 -1.3 1.310 14 4.610 13 1.8 279.4 -2.1 9.610 13 2.710 13 2.8 330.2 -3.8 6.910 13 1.810 13 3.0 5.2 Plasma Environment The plasma environment was measured in a region 15.2 cm in the axial direction by 5.08 cm in the vertical direction, 1.91 cm above the sample plate with an electrostatic Langmuir probe and Faraday probe. The scanning area was divided into 91 measurement points with a spatial resolution of 1.27 cm by 1.27 cm. An emissive probe scanned a region 15.2 cm in the axial direction by 6.78 cm in the vertical direction, 0.40 cm above the sample plate to capture the ambient plasma potential and sheath potential above the sample. The scanning area was divided into 286 measurement points with a spatial resolution of 0.06 cm by 1.27 cm within the sheath and 1.27 cm by 1.27 cm above the sheath. Figure 5.3 shows the measured contour plots of the plasma potential, p , current density, J i , ion density, n i , electron density, n e , space charge, n i n e , and electron temperature, T e . Note that the closest measurements to the target sample were made 19.1 mm above the surface, and the centerline of the plasma beam was 25.4 mm above the surface. The measured potential at the source center line above the leading edge of the plate (x = 179 mm, z = 25.4 mm) was used as the reference ambient potential. Table 5.1 shows the average measured plasma potential, ion density, electron density, and electron temperature at various x-positions along the beam centerline. The Debye length in the plasma environment was approximately D = 2 mm. Using the average ion density, electron density, and electron temperature from Table 5.1 and applying the standard 1-D space-charge limited current collection analysis for cold ions andthermalelectrons(Eq.3.2),theaveragecleanGore-Texsurfacepotentialis s '12:5V with respect to the ambient reference, and the sheath thickness is d sh ' 4 mm above the Gore-Tex surface. 5.3 Dusty Gore-Tex Samples 84 (a) p , plasma potential (b) J i , ion current density (c) n i , ion density (d) n e , electron density (e) n i n e , space charge (f) Te, electron temperature Figure 5.3: Plasma environment above the sample plate. The centerline of the plasma beam is at z = 19.1 mm. The reference potential is the potential at x = 179 mm, z = 19.1 mm. The potential values plotted are offset by the reference potential 5.3 Dusty Gore-Tex Samples Two different sample plates were made to measure clean and dusty Gore-Tex sample surface potentials and dust surface potentials. For each sample plate, nine 22-gauge PTFE high- temperature stranded wires were inserted through the bottom of the aluminum plate and arranged as shown in Figure 5.4. The h1 plate contained wires placed 0.08 cm above the aluminumplatesurface, whichalignedwiththesurfaceoftheGore-Texsampleandmeasured the potential of both clean and dusty Gore-Tex (Fig. 5.5(a) & 5.5(b), respectively). The h2 plate contained wires placed 0.32 cm above the aluminum plate surface, which aligned with the surface of a 0.24 cm thick dust layer and measured the potential of the dust layer (Fig. 5.5(c)). Each wire represents the potential of a 2.54 cm 2.54 cm square region on 5.4 Dusty Gore-Tex Charging 85 (a) Clean GoreTex sample (b) Clean GoreTex sample (c) Dusty GoreTex sample (d) Dusty GoreTex sample Figure 5.4: Clean and dusty Gore-Tex sample setups the Gore-Tex surface. Though the potential in each square region is unlikely to be uniform due to the ambient plasma environment, this coarse measurement is sufficient for this study. To determine how the dust accumulation affected the charging of an insulating material, the surface potentials of a clean Gore-Tex sample were measured and compared with that of a dusty Gore-Tex sample. The dusty Gore-Tex sample was created by placing a 0.24 cm thick layer of dust onto the centerline of the Gore-Tex sample (Fig. 5.4(d)). 5.4 Dusty Gore-Tex Charging Table 5.2 presents the measured floating potentials of clean and dusty Gore-Tex surfaces, as well as the measured dust surface potentials. The error bars on the calculated clean Gore- Tex surface potentials were determined by the accuracy of the emissive probe (10%), and the measured floating potential error bars were calculated by dividing the standard deviation 5.4 Dusty Gore-Tex Charging 86 (a) Clean GoreTex potential measurement setup (b) Dusty GoreTex potential measurement setup (c) Dust potential measurement setup Figure 5.5: h1 and h2 plate setups for clean and dusty samples of the entire wires’ sample by the square root of the sample size. Each test setup was run four times, and each wire was measured 12 times per run for a total of 48 measurements per test setup. The average clean Gore-Tex potential was -12.2 V, which matches closely with the average calculated potential of -12.5 V. Figure 5.6 visualizes the data in Table 5.2 in the same layout as Figure 5.4. The white squares represent Gore-Tex surface potentials, and the brown squares represent dust surface potentials. Figure 5.6(a) shows the measured surface potentials of a clean Gore-Tex surface. Figure 5.6(b) shows the measured surface potentials of clean Gore-Tex (D, E, F, K, M, and N) but G, H, and J are Gore-Tex surface potentials beneath a 0.24 thick dust layer. Figure 5.6(c) also shows the measured surface potentials of clean Gore-Tex (D, E, F, K, M, and N), but G, H, and J represent the dust layer surface potentials. Though a dust layer is present in Figure 5.6(c), the potentials at G, H, J, K, M, and N in Figures 5.6(a) and 5.6(c) look quite similar. This is to be expected because the charging at those locations should follow the space-charge limited current collection analysis, and the dust layer at G, H, and J should charge similarly to clean Gore-Tex at the same location. 5.4 Dusty Gore-Tex Charging 87 Table 5.2: Gore-Tex and dust surface potentials Position Calculated Clean Measured Clean Measured Dusty Measured Gore-Tex [V] Gore-Tex [V] Gore-Tex [V] Dust [V] D -13.51:5 -12.730:05 -13.680:13 - E -13.11:4 -13.180:07 -15.190:06 - F -13.71:4 -12.640:12 -14.500:01 - G -11.81:3 -10.500:04 -11.890:01 -10.500:10 H -13.41:3 -14.110:01 -16.580:02 -14.490:10 J -11.91:5 -10.260:04 -12.300:11 -10.730:12 K -11.01:4 -10.150:04 -9.910:07 - M -13.21:3 -15.310:01 -15.220:04 - N -11.41:4 -10.620:10 -10.710:06 - (a) Clean GoreTex surface potentials (b) Dusty GoreTex surface potentials (c) Dust layer & GoreTex surface potential Figure 5.6: Clean and dusty Gore-Tex surface potentials 5.5 Differential Charging 88 However, D, E, and F in Figures 5.6(b) and 5.6(c) are1.57 V more negative than in Figure 5.6(a). This is due to the plasma wake generated at the trailing edge of the dust layer, which drives the potentials of the surfaces behind the trailing edge more negatively, as was discussed in Chapter 4. The Gore-Tex surface potentials below the dust layer are 1.97 V more negative than that of a clean Gore-Tex surface. This supports the findings in Chapter 3 that the potential of a dust covered surface is driven negatively due to the capacitance formed by the dust layer, assuming the dust layer is a parallel plate capacitor. Because the dust surface potential is determined by current collection from the ambient plasma, the charge stored in the dust layer drives the potential of the Gore-Tex surface below it more negative. Assuming the dust layer collects 3.510 8 C/m 2 , as was found in Chapter 3, it is estimated that the dust layer will drive the potential of the surface beneath it 2.2 V more negative, which agrees well with what was measured. 5.5 Differential Charging Insulating materials will experience differential charging due to their inability to distribute charge evenly across their surfaces. Figure 5.6 clearly illustrates this phenomenon. The clean Gore-Tex sample sees a maximum potential difference of 5.05 V across its surface, whereas the dusty Gore-Tex sample sees a maximum potential difference of 6.67 V. Results show that differential charging increases on an insulating material surface partially covered by dust due to effects from the plasma wake and dust layer capacitance. The differential charging seen on the dust surface (G, H, and J in Fig. 5.6(c)) is directly translated onto the Gore-Tex beneath it (G, H, and J in Fig. 5.6(b)). Given that dust may accumulate at different thicknesses at different locations on an insulating material, variations of dust thickness and dust coverage will also increase differential charging. Thicker dust layers will create larger plasma wake regions behind them, which will generate surfaces with more negative potentials, and it will also drive the potential of the insulating surface beneath it more negatively than a thinner dustlayer. Itwillbeparticularlyimportant tostudythechargingatthe dustlayer-insulating material junction, where the largest electric fields will develop due to the transition in surface 5.6 Summary and Conclusion 89 charging processes. On the front edge of the insulating surface (between K-M-N and G-H- J), surface charging will transition from ambient current balance collection to dust layer capacitor controlled charging. On the back edge of the insulating surface (between G-H-J and D-E-F), surface charging will transition from dust layer capacitor controlled charging to plasma wake charging. Both regions will see an increase in differential charging, which will increase the probability for arcing and discharge. 5.6 Summary and Conclusion A laboratory study was carried out to measure the charging of an insulating material, Gore- Tex, partially covered by a layer of lunar dust simulant, JSC-1A, in a collisionless, mesother- mal plasma. The floating potentials of the dust layer and of the insulating material beneath the dust layer were obtained utilizing the embedded-wire method. The experimental results show that differential charging occured on a clean Gore-Tex surface, as expected, but when a dust layer was introduced, the magnitude of differential charging on the insulating surface increased. The dust layer drove the potential of the insulating surface below it more nega- tively, as was found in Chapter 3, and the plasma wake region created behind the dust layer drove the potential of the insulating surface behind the dust layer more negatively, as was found in Chapter 4. Given that dust accumulation on an astronaut’s space suit will likely not be uniform, differential charging on a dusty space suit will be larger than that of a clean space suit, which increases the probability for arcing or discharge. 90 Chapter 6: Dusty Surface Charging Due to Ultraviolet Radiation On the lunar dayside surface, photoelectrons emitted by UV-illuminated surfaces dominate current collection and charge surfaces positively [5]. A clean conducting surface will charge to a positive potential due to the emission of photoelectrons (see Fig. 6.1(a), where ph is photoelectron flux). Because lunar dust and conducting metals have different material properties, however, dust coverage will change the charging process of a conducting surface illuminated by UV radiation. In this situation, the floating potential of the surface will be determined by both the current collection condition and the material properties of both the dust layer and conducting surface (Fig. 6.1(b)). When insulating materials, such as those found on astronauts’ space suits, are exposed to UV-illumination, those surfaces will also charge slightly positive. However, due to the insulating material’s inability of distributing the charge evenly throughout its surface, its surface will charge differentially (Fig. 6.1(c)). When the insulating material is in a dusty environment and covered by dust, it, too, will experience an altered charging process (Fig. 6.1(d)). Currently,thereisverylimitedknowledgeonhowadustyconductingoradustyinsulating surface charges under UV-illumination. This chapter studies the effects of JSC-1A lunar regolith coverage on the charging of both an aluminum plate and a Gore-Tex surface due to UV-illumination. 6.1 Experimental Setup Two commercial germicidal 25 W UV-C low-pressure bulbs with quartz sleeves were used to generate photons with energies of 6.7 eV, as was discussed in Chapter 2. The target samples were an electrically floating 15.2 cm 10.2 cm aluminum plate and a 7.6 cm 10.2 cm piece of Gore-Tex. They were placed 7.6 cm below the two UV sources, which hung from the center of the vacuum chamber (Fig. 6.2). Each sample was tested 6.1 Experimental Setup 91 (a) Clean conducting surface (b) Dusty conducting surface (c) Clean insulating surface (d) Dusty insulating surface Figure 6.1: Surface charging by UV illumination individually. Because surfaces with work functions lower than the emitted photon energies will generate photoelectrons, it was necessary to prevent “contaminating" photoelectrons emitted by surfaces other than the target sample, such as the stainless steel vacuum chamber walls (WF 4.4), from interfering with the target sample. A stainless steel grid with 76% transmission was used to surround the target sample on four sides, leaving the back side open for probes to enter, similarly to that of Dove et al [97], and the grid was connected to an external power source so that it could be biased to repel “contaminating" photoelectrons. To determine the appropriate grid bias to prevent “contaminating" photoelectrons from interacting with the target samples, a clean aluminum plate was utilized as a Langmuir probe. The resulting I-V curve contained the information necessary to calculate the pho- toelectron temperature, T ph , and photoelectron density, n ph , of the environment, as well as the floating potential, f , of the aluminum plate. At bias voltages more negative than the local plasma potential, V b < p , emitted photoelectrons are repelled and generate a positive 6.1 Experimental Setup 92 Figure 6.2: A schematic showing the experimental setup of the UV source and target sample plate current. At bias voltages more positive than the local plasma potential, V b > p , both emit- ted photoelectrons and ambient electrons are collected by the plate, generating a negative current. Figure 6.3 shows the measured I-V curves for a clean aluminum plate for multiple cage biases. The aluminum plate bias was swept from -40 V to +40 V in increments of 2 V. Results show that the current collected by the aluminum plate at large negative voltages, regardless of the grid bias, was constant. This represents the photoelectron current emitted by the aluminum plate; because the plate was biased more negatively than the biased grid, electrostatic forces pointed outward from the plate surface towards the biased grid, prevent- ing emitted photoelectrons from returning to the plate. At large positive voltages, however, the collected electron current magnitude was far greater than the emitted photoelectron current magnitude. This was due to “contaminating" photoelectron collection. Introducing the biased grid reduced “contaminating" photoelectron effects; at small negative grid bias voltages, some “contaminating" photoelectrons had enough energy to pass through the grid and to be collected by the plate, but at larger negative biases, the grid prevented those photoelectrons from entering the sample plate’s environment. Figure 6.3 shows that the col- lected electron current was similar to the emitted photoelectron current when the grid was biased to -40 V. To further reduce vacuum chamber effects and preserve the target sample photoelectron environment, a positively biased 10.2 cm 10.2 cm aluminum “collecting” plate was hung 1.3 cm above the grid and between the two UV-C bulbs. When the grid 6.2 Photoelectron Environment 93 Figure 6.3: Aluminum plate I-V curves as a function of cage bias was biased to -20 V and the collecting plate was biased to +30 V, the electron current at large positive biases decreased significantly (Fig. 6.4), confirming that effects from photo- electrons emitted by other surfaces had been reduced, and the plasma environment above the aluminum plate was mainly generated by photoemission from the plate. It should be noted that the photoelectron current decreased by a factor of two when the collecting plate was introduced. This occurred because the placement of the collecting plate reduced photon flux to the target sample. All dusty-surface charging experiments utilized the -20 V grid and +30 V collecting plate setup. 6.2 Photoelectron Environment Before performing charging experiments, it was necessary to characterize the photoelectrons emitted by aluminum, JSC-1A, and Gore-Tex. The maximum photoelectron energy, T ph , emitted from a clean surface is determined by the difference in photon energy and the work function of the material: 6.2 Photoelectron Environment 94 Figure 6.4: Aluminum plate I-V curves with cage biased at -20 V and collecting plate biased at +30 V Table 6.1: Maximum photoelectron temperature Material Work Function [eV] Maximum T ph [eV] Aluminum 4.2 2.5 Gore-Tex 5.8 0.9 JSC-1A 5.8 0.9 T ph =EW (6.1) where E is the photon energy, and W is the work function of the material. Due to contamination of sample surfaces, such as from oxidation, for example, the actual photo- electrons generated from sample surfaces may have energies either smaller or larger than the theoretical maximum. Table 6.1 presents the maximum photoelectron energy for clean aluminum, clean Gore-Tex, and clean JSC-1A. An aluminum plate with no dust coverage (Fig. 6.5(a)), a full dust layer sample (Fig. 6.5(f)), and a Gore-Tex sample with no dust coverage (Fig. 6.6(a)) were each charged by UV-illumination to determine the energy of photoelectrons generated by each surface. 6.2 Photoelectron Environment 95 Table 6.2: Ambient plasma parameters 0.8 cm above target sample surface determined by Langmuir probe sweeps Sample f [V] T ph [eV] n ph [m 3 ] Y Aluminum 2.21 2.4 1.410 7 310 2 [98] JSC-1A 0.60 1.2 1.510 5 510 4 [99] Gore-Tex 0.53 1.0 1.310 3 110 5 [100] Utilizing a clean aluminum plate as a Langmuir probe, as described previously in Chap- ter 6.1, 2.5 eV photoelectrons were measured for an aluminum plate. From Figure 6.4, it was determined that the aluminum plate collected a photoelectron current, I ph , of 9.4310 9 A, which corresponded to a current density, J ph , of 0.608 A=m 2 . For a negatively biased surface, current collection by photoelectron emission can be modeled by Equation 6.2: J ph =en ph r KT ph 2m e ;V b < p (6.2) This resulted in a measured photoelectron density of 1:4 10 7 =m 3 . An electrostatic Langmuir probe was also used to measure the photoelectron temperature and density 0.8 cm above the aluminum plate surface (Table 6.2), validating the results. Using the embedded- wire method, the clean aluminum plate was found to float at +2.2 V with respect to the biased grid. The measurement shown in Figure 6.4 measured the aluminum plate’s floating potential to be +2.4 V with respect to the biased grid, also confirming this result. A surface charged only by UV-illumination will float to a positive potential that returns all emitted photoelectrons. Therefore, a surface will float to a potential near the emitted photoelectron temperature. Because the aluminum-emitted photoelectrons had energies of 2.4 eV, it was expected that the aluminum plate would float +2.4 V with respect to the biased grid. Hence, the measured +2.2 V also agrees well with theory. Given that JSC-1A and Gore-Tex could not be electrically biased, a Langmuir probe was used to measure their photoelectron temperature and density. Good agreement between the aluminum plate measurements and Langmuir probe measurements provided confidence that the Langmuir probe could characterize the photoelectron environment accurately. Table 6.2 6.3 Dusty-Surface Samples 96 presents the measured surface floating potential, f , photoelectron temperature, T ph , pho- toelectron density, n ph , and photoelectric yield, Y, for all three samples. 6.3 Dusty-Surface Samples Todeterminetheeffectsofdustcoverageonthechargingofaconductingplateilluminatedby UV radiation, six different dusty-sample plates were used (Fig. 6.5). For each dusty-sample setup, the dust layers were 0.16 cm thick. To determine the effects of dust coverage on the charging of an insulating surface illuminated by UV radiation, the potentials of “clean” Gore- Tex (Fig. 6.6(a)) and “dusty” Gore-Tex (Fig. 6.6(b)) surfaces were compared. The “dusty" Gore-Tex surface had a dust layer 0.24 cm thick. (a) Clean plate (b) 1 sq (c) 7 sq 3 0 1 0 3 (d) 8 sq 0 3 2 3 0 (e) 14 sq (f) Full dust coverage Figure 6.5: Dusty-sample setups 6.4 Photoelectron Charging of Dusty Conducting-Surface 97 (a) Clean suit (b) Dusty suit Figure 6.6: Gore-Tex sample setups 6.4 Photoelectron Charging of Dusty Conducting- Surface Surface potential contours of the dusty aluminum plate samples (Figure 6.5) are presented in Figure 6.7. White squares represent exposed aluminum plate surfaces, and brown squares representdust-coveredareas. Eachsquarerepresentsthepotentialmeasuredbyanindividual wire. Results show that as dust coverage increased, the conducting plate’s floating potential and the dust surface potential both became less positive. This is largely due to the difference betweenthetwomaterials’workfunctions. Becausealuminunhasalowerworkfunctionthan JSC-1A, it generates “higher” energy photoelectrons and forces the aluminum plate to charge more positively than JSC-1A (Table 6.2) to satisfy the current balance requirement. When dust was introduced onto the aluminum plate’s surface, the electrostatic forces pointed from the dust surface towards the aluminum plate, causing the dust-emitted photoelectrons to be collected by the aluminum plate and allowing the aluminum plate to charge less positively. In addition, the dust layer acted as a parallel plate capacitor, which drove the potential of the aluminum plate beneath it less positively. As more dust was introduced onto the surface 6.4 Photoelectron Charging of Dusty Conducting-Surface 98 Table 6.3: Average measured dusty sample surface potentials and charge density Sample plate [V] dust [V] [e=m 2 ] Clean plate 2.21 - - 1 sq. 2.17 0.78 2.0210 11 7 sq. 3-0-1-0-3 2.08 0.75 1.9410 11 8 sq. 0-3-2-0-3 2.11 0.74 1.9910 11 14 sq. 1.91 0.61 1.9010 11 Full dust -0.64 0.60 1.8110 11 of the aluminum plate, more “low-energy” photoelectrons were generated by the dust surface, reducing the magnitude of the potential on both surfaces. When dust completely covered the aluminum plate, the aluminum plate’s floating potential was dependent on the charge stored on dust layer surface and on the dust layer capacitance. Table 6.3 shows the average measured aluminum plate and the dust surface potentials, as well as the calculated charge density, , collected by the dust surface in units of elementary charge per square meter. The results show that the charge density collected by the dust layer decreased as more dust was introduced onto aluminum plate surface. This confirms that as dust coverage increases, the dust layer charges less positively to satisfy the current balance requirement. Theexperimentalresultsdemonstratedthatdust-coveredconductingsurfacesilluminated by UV radiation will charge less positively than that of a clean conducting surface. Addi- tionally, the difference between aluminum and JSC-1A’s surface potentials ( 1.7 V) for a dusty plate sample was approximately the difference between their work functions, which is approximately 1.5 eV. Therefore, a dust-covered conducting surface charged purely by UV radiation will not exhibit large differential charging. However, a dust-covered conducting surface that is charged by other current sources, such as from the solar wind, in addition to photoelectron current may exhibit larger differential charging depending on the geometry and orientation of the surface, as was described in Chapter 1.1.2. 6.4 Photoelectron Charging of Dusty Conducting-Surface 99 (a) Clean plate (b) 1 sq (c) 7 sq 3 0 1 0 3 (d) 8 sq 0 3 2 3 0 (e) 14 sq (f) Full dust coverage Figure 6.7: Dusty aluminum plate potential contours 6.5 Photoelectron Charging of Dusty Insulating Surface 100 Table 6.4: Average measured Gore-Tex and dust layer surface potentials Sample clean [V] dustcovered [V] dust [V] Clean Gore-Tex 0.53 - - Dusty Gore-Tex 0.44 -0.39 0.48 6.5 Photoelectron Charging of Dusty Insulating Surface Figure 6.8 presents the measured clean and dusty Gore-Tex surface potentials, as well as the dust surface potential. White squares represent Gore-Tex surfaces, and brown squares represent dust surfaces. Comparing Figure 6.8(a) with Figure 6.8(c), there was minimal differential charging on the surface of clean and dusty Gore-Tex samples. Because Gore-Tex and JSC-1A have similar work functions (Table 6.1), they both charge to similar potentials to balance the emitted photoelectron current with collected electron current. The potential of a clean Gore-Tex surface was slightly more positive than that of the clean surfaces for a dust-covered Gore-Tex surface. This was due to the difference in photoelectric yield between Gore-TexandJSC-1A.JSC-1AhasahigherphotoelectricyieldthanGore-Tex, sowhen JSC- 1A was introduced, increased photoelectron density increased electron current collection and allowed the Gore-Tex surface to charge less positively. Figure 6.8 also illustrates the differential charging between dust-covered and clean Gore- Tex surfaces. Clean Gore-Tex surfaces will charge according to current balance, but dust- covered Gore-Tex surface potentials are determined by the dust layer capacitance above it. Table 6.4 displays the average measured surface potentials, where clean are potentials of Gore-Tex surfaces not covered by dust, dustcovered are potentials of Gore-Tex surfaces covered by dust, and dust are dust surface potentials. The results show that the dust layer acted as a parallel plate capacitor and drove the Gore-Tex surface potential beneath it 0.77 V more negatively. These results indicate that differential charging will increase on insulating surfaces cov- ered by dust. The boundaries between clean and dust-covered surfaces, in particular, are prone to larger differential charging because the surface potentials are driven by two differ- ent processes. Clean surfaces are driven by current balance, and dust-covered surfaces are driven by dust layer capacitance. Because the difference in work function between JSC-1A 6.5 Photoelectron Charging of Dusty Insulating Surface 101 (a) Clean GoreTex (b) Dusty GoreTex (c) Dust + GoreTex Figure 6.8: Clean and dusty Gore-Tex potential contours and Gore-Tex is minimal, however, the magnitude of differential charging due purely to UV-illumination is not expected to pose any significant problems. Similarly to a conducting surface, on the other hand, an insulating surface exposed to other current sources in addition to photoelectron currents may exhibit larger differential charging depending on orientation, geometry, and material properties. 6.6 Electrostatic Potential Threshold Between JSC-1A and Aluminum 102 6.6 Electrostatic Potential Threshold Between JSC-1A and Aluminum The results obtained in Chapter 4 showed that patches of dust accumulation on a conducting plate can drive the potential of the conducting plate more negatively than that of a con- ducting plate completely covered by a uniform layer of dust due to the plasma wake region generated at the trailing edge of the dust patches. Given that the lunar surface is exposed to very rarefied plasma in the magnetotail lobes and more energetic and turbulent plasma in the plasma sheet and magnetosheath, it is likely that large differential charging will occur on a dusty conducting surface. Thus, it is crucial to determine the potential threshold for electrostatic discharge to occur between dust and a conducting surface. In this chapter, we describe an attempt to determine the potential threshold for electrostatic discharge (ESD) to occur between JSC-1A and aluminum. 6.6.1 Experimental Setup The ESD potential threshold test utilized the experimental setup described in Chapter 6.1, but the biased grid was removed to ensure that the discharge only occurred between the aluminumplateandtheJSC-1Adustsurface. Thepositively-biasedcollectingplateremained in the test setup. The 8 sq. 0-3-2-3-0 dusty sample without embedded wires (Fig. 6.9(a)) and an aluminum plate randomly covered with dust (Fig. 6.9(b)) were used for this experiment. The aluminum plate was electrically connected to a high-voltage (HV) power supply outside of the vacuum chamber, which could bias the aluminum plate from 0 to -4000 V. It has been found that electrical breakdown of dielectrics begins when the electric field is on the order of 10 8 V/m [101]. Because 4000 V was the largest bias potential the HV power supply could provide, three different dust thicknesses (1.6 mm, 0.1 mm, and 0.025 mm) were used to increase the electric field between the dust surface and aluminum plate. To detect ESD events, a Tektronix TDS640 oscilloscope was connected in line with the aluminum plate and HV power supply (Fig. 6.10). A 10000:1 voltage probe was wired 6.6 Electrostatic Potential Threshold Between JSC-1A and Aluminum 103 (a) 8 sq 0 3 2 3 0 (b) Random coverage Figure 6.9: Electrostatic discharge dusty sample setup Figure 6.10: Electrostatic discharge test wiring setup between the oscilloscope and aluminum plate, and a 1 M resistor was placed between the HV power supply and dusty plate sample to protect the power supply. During ESD testing, the aluminum plate bias was increased incrementally in steps of 500 V from 0 to -4000 V, and at each step, the dusty sample sat under UV-illumination for 900 seconds. To ensure the test setup could detect ESD events, a clean aluminum plate was biased to -4000 V, and a grounded rod was used to short the setup. Figure 6.11 shows the output from the oscilloscope due to the short and provided confidence that the test setup could detect ESD events. 6.6 Electrostatic Potential Threshold Between JSC-1A and Aluminum 104 Figure 6.11: Example of electrostatic discharge event detected by oscilloscope Table 6.5: Average dust layer potentials with biased aluminum plate Plate Bias [V] dust [V] 0 3.12 -30 3.21 -60 3.57 -100 4.03 6.6.2 Dusty-Surface Charging with Negatively Biased Aluminum Plate Before performing ESD testing, it was necessary to determine how the dust surface charged underUV-illuminationwhenthealuminumplatewasbiasednegatively. InChapter6, results showed that the 8 sq. 0-3-2-3-0 dust layer charged to 0.74 V when the aluminum plate was floating and with a -20 V biased grid above it. With a negatively-biased aluminum plate but without a biased grid, the dust surface charged to larger positive potentials. This was likely due to the stronger electric field that pointed from the dusty sample towards the collecting plate (+30 V). Table 6.5 shows that as the aluminum plate’s bias voltage increased in magnitude, the dust surface charged more positively. Increasing the aluminum plate bias increased the electrostatic force between the aluminum plate and collecting plate, forcing the dust surface to charge more positively to collect emitted photoelectrons. The results 6.7 Summary and Conclusion 105 Table 6.6: Approximate electric field strength as function of dust thickness Dust Thickness [mm] E [V=m] 1.6 2.510 6 0.1 4.010 7 0.025 1.610 8 obtained in Table 6.5 provided confidence that the dust surface would charge positively when the aluminum plate was biased negatively to kV magnitudes. 6.6.3 Electrostatic Discharge Testing The approximate electric field strength between the dust surface and aluminum plate as a function of dust thickness is shown in Table 6.6. The largest electric field was generated with the thinnest dust layer (0.025 mm) and largest plate bias voltage (-4000 V), but no ESD events were detected during the entire experiment. It was determined that a higher electric field would be necessary to generate breakdown. Given that an electric field on the order of 10 8 V/m does not generate ESD between JSC-1A and aluminum, dusty surfaces charged in a low-energy plasma environment, such as that generated in this dissertation, will likely not be prone to ESD. 6.7 Summary and Conclusion A laboratory study was carried out to measure the charging of dust-covered conducting and insulating surfaces by UV-illumination. The floating potentials of the dust layer, the conducting plate, and the insulating surface beneath the dust layer were obtained utilizing the embedded-wire method. The results showed that dust-covered surfaces charged only by UV radiation will display differential charging on the order of the difference between work functions, which is generally a few eV’s, and will not be prone to arcing or discharge. Dusty surfaces exposed to other current sources in addition to photoemission, however, may generate large differential charging depending on geometry, orientation, and material properties. 6.7 Summary and Conclusion 106 In addition, no ESD events were detected when the electric field between the dust surface and aluminum plate was1.610 8 V/m. This provides confidence that ESD is not a concern in low-energy plasma environments, such as those generated in this dissertation. It will still benecessarytodeterminethepotentialthresholdforbreakdowntooccurforsituationswhere dusty surfaces charge in areas prone to high-energy particles and radiation, such as those found in permanently shadowed regions on the Moon or in Earth’s magnetotail. 107 Chapter 7: Plasma Charging of Dusty Surfaces During Multibody-Plasma Interactions When an astronaut is performing an extravehicular activity (EVA), such as walking by a lunar boulder or coming in close contact to a lunar rover, multibody-plasma interactions may occur. Multibody-plasma interactions (MPI) refer to the situation when two or more charged bodies in a plasma simultaneously interact with the plasma and with each other [102]. On the lunar surface, EVA occurs in a dusty plasma environment. This chapter studies dust coverage effects on surface charging under the multibody-plasma interaction situation. A clean insulating surface’s charge is dependent on its local ambient environment (Fig. 7.1(a)). When insulating materials are in close proximity to other surfaces, how- ever, the charging process becomes dependent on both the ambient plasma environment and on the other surface’s interactions with the plasma. Whether the other surface is a clean conducting surface (Fig. 7.1(b)) or a dusty dielectric surface (Fig. 7.1(c)) will have differ- ent effects. This chapter presents an experimental investigation on dusty surface charging under MPI by looking at the charging of an “astronaut’s arm" in close proximity to a clean aluminum plate or a layer of JSC-1A in a mesothermal plasma of cold ions and thermal electrons at 0 angle of attack. 7.1 Experimental Setup The simulated astronaut arm used for this experimental investigation was a 3.8 cm 2.5 cm 10.2 cm aluminum rectangular prism wrapped on all sides by Gore-Tex. An electrically floating 10.2 cm 15.2 cm aluminum plate was used to simulate a spacecraft’s surface, and a 0.16 cm thick dust layer partially covering an aluminum plate was used to simulate a dusty surface. The astronaut’s arm was placed 20.3 cm downstream of the plasma source 7.1 Experimental Setup 108 (a) Clean insulating surface (b) Clean insulating and clean conducting surface (c) Clean insulating and dusty surface Figure 7.1: Plasma flux to insulating surface near clean, conducting and dusty surface and along the centerline of the plasma source (Fig. 7.2). The aluminum plate was placed 17.8 cm downstream of the plasma source and 0.6 cm below the astronaut’s arm, creating a narrow channel between the -z face of the astronaut’s arm and the plate surface (Fig. 7.3). This is the area of interest. The astronaut’s arm and the plate orientation were both parallel to the plasma beam direction. Because the cold ion plasma beam had about a 0 angle of attack with respect to the target sample, the plasma current collected at the plate surface was approximately that of cold ions and thermal electrons. The embedded-wire method was used to measure the floating potential of the clean aluminum plate and the surface potentials of the astronaut’s arm and the dust layer. Three wires were placed on each of the four outer faces of the astronaut arm (+y, +z, -y, -z face) and arranged as shown in Figure 7.4. Using the plasma source exit as a reference point, the wires on each face were placed at the x-positions shown in Table 7.1. This wire arrangement allowed for the measurement of the entire astronaut arm’s surface potential. Embedded 7.1 Experimental Setup 109 Figure 7.2: Schematic showing the layout of the plasma source, astronaut arm, and dusty/- conducting surface (a) Astronautarmincloseproximity toa clean aluminum plate (b) Astronaut arm in close proximity to a dusty surface Figure 7.3: Setup of astronaut arm near clean conducting and dusty surface wires were also placed at the same location with respect to source exit on the dusty plate to measure dust potentials directly opposite the measured Gore-Tex surface potentials. The surface potentials on the -z face were of most interest. Surface potentials on the -y, +y, and +z faces charge according to standard space-charge limited current collection from the ambient plasma, but ambient plasma conditions will change in the channel formed between the astronaut arm and dusty surface. Table 7.1: Wire position distance from plasma source exit Position Distance [cm] x1 20.3 x2 22.9 x3 25.4 7.2 Plasma Environment 110 Figure 7.4: Astronaut arm and dust layer embedded wire setup Table 7.2: Average ambient plasma parameters at various x-positions along beam centerline x-position [mm] p [V] n i [m 3 ] n e [m 3 ] T e [eV] 177.8 0.0 1.610 14 5.410 13 1.8 228.6 -1.3 1.310 14 4.610 13 1.8 279.4 -2.1 9.610 13 2.710 13 2.8 330.2 -3.8 6.910 13 1.810 13 3.0 7.2 Plasma Environment Figure 7.5 shows the measured contour plots of the plasma potential, p , current density, J i , ion density, n i , electron density, n e , space charge, n i n e , and electron temperature, T e . Note that the closest measurements to the aluminum plate were made at 19.1 mm above the surface, and the centerline of the plasma beam was 25.4 mm above the surface. The measured plasma parameters at the source center line above the leading edge of the plate (x = 179 mm, z = 25.4 mm) were used as the reference plasma potential. Figure 7.6 presents the potential field around the astronaut arm in the Y-Z plane when it was inserted into the plasma field. Table 7.2 shows the average measured plasma potential, ion density, electron density, and electron temperature at various x-positions along the beam centerline. Because this experimental setup is identical to the one presented in Chapter 5, the average ambient plasma parameters are identical to those presented in Chapter 5. The Debye length in the plasma environment was approximately D = 2 mm. 7.2 Plasma Environment 111 (a) p , plasma potential (b) J i , ion current density (c) n i , ion density (d) n e , electron density (e) n i n e , space charge (f) Te, electron temperature Figure 7.5: Plasma environment above the sample plate. The centerline of the plasma beam is at z = 19.1 mm. The reference potential is the potential at x = 179 mm, z = 19.1 mm. The potential values plotted are offset by the reference potential Figure 7.6: Plasma potential in Y-Z plane around the astronaut’s arm 7.3 Astronaut-Arm Charging 112 Table 7.3: Average ambient plasma parameters at various x-positions along beam centerline of the three different plasma environments presented x-position [mm] p [V] n i [m 3 ] n e [m 3 ] T e [eV] 177.8 0.0 1.110 14 4.610 13 1.8 228.6 -1.7 8.610 13 3.310 13 2.0 279.4 -2.8 7.110 13 2.310 13 3.1 330.2 -4.0 5.810 13 1.710 13 2.6 A total of three different mesothermal plasma environments were presented in this disser- tation (Chapters 3 and 4, and Chapter 5 and this chapter presented idential environments). The average ambient plasma parameters of the three plasma environments at various x- positions are shown in Table 7.3. Figure 7.7 plots the average, minimum, and maximum plasma parameters at various x-positions along the beam centerline. Based upon the results in Table 7.3 and shown in Figure 7.7, the following trends have been identified: The plasma potential becomes monotonically more negative with increasing x-position. The electron temperature shows a more complicated variation with increasing x- position, and only at x = 279.4 mm was the electron temperature significantly higher than it was at any of the other three positions. The ion density decreases monotonically with increasing x-position. The electron density decreases monotonically with increasing x-position and is2x lower than the ion density at each x-position. 7.3 Astronaut-Arm Charging Using the average ion density, electron density, and electron temperature from Table 7.2 and applying the standard 1-D space-charge limited current collection analysis (Eq. 3.2), the average surface potentials along the +y, +z, and -y faces were s '12:5V with respect to the ambient potential. This agrees well with the average measured astronaut’s arm surface potentials along the +y, +z, and -y faces shown in Tables 7.4 and 7.5. For an astronaut’s 7.3 Astronaut-Arm Charging 113 (a) p , plasma potential (b) Te, electron temperature (c) n i , ion density (d) n e , electron density Figure 7.7: Average, minimum, and maximum ambient plasma parameters at various x- positions along beam centerline arm in close proximity to a clean aluminum plate, the average surface potential was -13.40 V; for an astronaut’s arm in close proximity to a dusty surface, the average surface potential was -13.29 V. Table 7.4 presents the measured surface potentials of the astronaut arm when set 0.6 cm from a clean aluminum plate. The standard errors presented in the table were calculated by dividing the standard deviation of the entire wires’ sample by the square root of the sample size. Each test setup was ran four times, and each wire was mesaured 12 times per run for a total of 48 measurements per test setup. The results show that the aluminum plate floated at -19.44 V, and the -z face of the astronaut’s arm charged from -21.68 V at x1 to -28.92 V 7.3 Astronaut-Arm Charging 114 Table 7.4: Measured astronaut arm surface potentials with clean aluminum plate [V] Plane x1 x2 x3 +y -13.890:04 -15.970:18 -16.580:09 +z -10.310:14 -8.210:05 -9.160:08 -y -14.550:10 -14.980:28 -16.960:06 -z -21.680:05 -24.520:01 -28.920:09 Clean Plate -19.440:05 -19.440:05 -19.440:05 Table 7.5: Measured astronaut arm surface potentials with dusty surface [V] Plane x1 x2 x3 +y -13.890:03 -14.790:05 -15.830:05 +z -10.090:07 -8.880:07 -11.250:14 -y -12.950:07 -14.980:29 -16.960:07 -z -9.900:08 -24.560:01 -31.880:07 JSC-1A -7.240:57 -18.400:26 -24.670:26 at x3, a 7.21 V potential difference. This generated an electrical field strength ranging from 373 V/m at x1 to 1580 V/m at x3. Table 7.5 presents the measured surface potentials of the astronaut’s arm when set 0.6 cm from a JSC-1A dust layer. The results show that both the dust layer and -z face of the astronaut’s arm displayed much greater differential charging across their surfaces compared to the astronaut’s arm when it was near an aluminum plate. Dust surfaces charged from -7.24 V at x1 to -24.67 V at x3, and the astronaut’s arm charged from -9.90 V at x1 to -31.88 V at x3, a 21.98 V potential difference. The electric field generated between the astronaut’s arm and the dusty surface ranged from 443 V/m to 1202 V/m. Comparing the floating potential of a clean aluminum plate without the presence of an astronaut’s arm (-12.5 V) with the floating potential of a clean aluminum plate with the presence of an astronaut’s arm (-19.4 V), it can be seen that presence of the astronaut’s arm causes the clean aluminum plate to charge more negatively. This is likely due to a large decrease in ion density as ions travel from the ambient plasma environment and into the narrow channel. More importantly, the electric field strength generated between the astronaut’s arm and the surface increased further downstream, regardless of whether the surface was conducting or dusty. This indicates that there will be a length threshold on an insulating object at 7.4 Summary and Conclusion 115 which the electric field will be strong enough to generate arcing or breakdown. Decreasing the distance between the astronaut’s arm and the surface will significantly increase electric field strength and increase the probability for electrostatic discharge. 7.4 Summary and Conclusion A laboratory study was carried out to measure the charging of an "astronaut’s arm" in close proximity to a conducting or dusty surface. The surface potential along the astronaut’s arm and dust surface, as well as the floating potential of the conducting surface, were obtained utilizingtheembedded-wiremethod. Theresultsshowedthatsignificantlygreaterdifferential charging occurs on an astronaut’s arm when it is in close proximity to a dusty surface, and the electric field strength between the astronaut arm and the dusty surface increases further downstream from the ram side. Depending on the length of the astronaut’s arm and the distance between the arm and the surface, breakdown or arcing may become a significant concern. This is relevant in situations where an astronaut must reach over a surface to operate a sensitive piece of equipment or to grab a tool. 116 Chapter 8: Summary and Conclusions Thenear-surfaceenvironmentattheMoonandmostasteroidsisadustyplasma. Thisdisser- tation experimentally investigated the effects of dust coverage on the charging of conducting and insulating surfaces in a dusty plasma environment.. 8.1 Conclusions and Contributions This research set out to provide answers and insights to the following questions: How does the charge on a dust grain in a “dusty surface” charge compare to that of a single, isolated dust grain? How does the charging of a dust-covered conducting surface differ from that of a clean conducting surface? How does the charging of a dust-covered insulating surface differ from that of a clean insulating surface? How do multibody-plasma interactions differ for a clean and a dusty surface? All of these questions were addressed in the context of the proceeding four chapters. The conclusions reached from the various experimental investigations and their resulting implications are discussed in this section. 8.1.1 “Dusty Surface" vs. “Dust-in-Plasma" Previously, studiesondustchargingonlyconsideredisolateddustgrains. Inthisdissertation, the charging of a dusty surface was measured and presented. Chapter 3 presented the first laboratory measurement of dusty-surface charging in plasma. Embedded wires were placed 8.1 Conclusions and Contributions 117 at various heights with respect to the dust layer and electrically connected to a measurement plate placed outside of the vacuum chamber, where a Trek non-contacting electrostatic volt- meter measured the floating potential of the measurement plate. The experimental results obtained in Chapter 3 validated the embedded-wire method and provided confidence that it could be used to measure dust surface and insulating surface potentials. Results on “dusty surface” charging found that the charge of a single, isolated dust grain is about three orders of magnitude larger than that of a charge per dust grain on a dusty surface, whichisingoodagreementwiththeory. Thecapacitanceofadustysurfaceisgreater than the capacitance of an individual dust grain. Thus, dust grains on a dusty surface do not need to accumulate as much charge to charge to the same potential as that of an individual dust grain. 8.1.2 Conducting Surfaces: Clean vs. Dusty The charging of a clean conducting surface was compared to that of a dusty conducting surface in a mesothermal plasma and under UV illumination. Chapter 4 presented an investigation on how the dust coverage quantity and the dust coveragepatternaffectedthechargingofaconductingplateinamesothermalplasma. Exper- imental results were compared with numerical simulations, and the investigation found that dust accumulation drives the potential of the conducting surface beneath it more negatively by two different methods. First, the dust layer can be modeled as a parallel plate capacitor. The charge stored on the dust layer surface is controlled by the current balance condition, and the capacitance of the dust layer, which is dependent on its thickness and its surface area, drives the potential of the conducting surface beneath it more negatively. Secondly, dust accumulation creates plasma wake regions at the trailing edge of dust layers due to expansion of the mesothermal plasma. This increases the electron current collected by the conducting plate, driving the potential of the conducting plate even more negatively. When a conducting surface is completely covered by dust, its floating potential is dependent on the dust layer surface charge, but when there are patches of dust accumulation, the potential of the conducting surface is dependent on both the dust layer capacitance and the generated 8.1 Conclusions and Contributions 118 plasma wake regions. Given that dust accumulation will likely be non-uniform in a dusty environment, conducting surfaces partially covered by dust may be prone to large differential charging between the dust layer and conducting surfaces. Chapter 6 presented the charging of both a clean and a dusty conducting surface under UVillumination. Resultsfoundthatdifferentialchargingwouldlikelynotbeamajorconcern for a conducting surface charged only by photoemission. The potential difference between the conducting surface and dust layer is driven by the difference in work functions between the materials, which is generally a few eV’s. The potential of a conducting plate completely covered by dust and exposed to UV illumination, is dependent on the capacitance of the dust layer, as was discussed previously. However, a dusty conducting surface exposed to other current sources in addition to photoemission may experience considerable differential charging depending on its geometry, orientation, and material properties, as was discussed in Chapter 1.1.2. 8.1.3 Insulating Surfaces: Clean vs. Dusty Chapter 5 presented the charging of clean and dusty Gore-Tex in a mesothermal plasma. Clean Gore-Tex surfaces displayed less differential charging than dusty Gore-Tex surfaces because their surface potentials were determined by local current collection. When a dust layer was introduced onto the Gore-Tex surface, the differential charging across the Gore- Tex surface increased due to effects from the generated plasma wake region and the dust layer capacitance. Given that dust may accumulate at different thicknesses at different locations on an insulating material, variations on dust thickness and dust coverage will increase differential charging. Thicker dust layers will drive the potential of the surface beneath it more negative and also create larger plasma wake regions, which charge surfaces behind it more negatively. It will be particularly important to understand the charging at the dust layer-insulating material junction, where the largest electric fields will develop due to the transition in surface charging processes. Chapter 6 presented the charging of both clean and dusty Gore-Tex under UV illumina- tion. Similar to results obtained for a conducting surface, differential charging will likely not 8.1 Conclusions and Contributions 119 pose a concern for dusty Gore-Tex surfaces charged only by photoemission because the differ- ence between work functions is minimal. Additionally, the charge stored on the dust surface due to photoemission is extremely small, creating little differential charging between clean Gore-Tex surfaces and dust-covered Gore-Tex surfaces. A dusty insulating surface exposed to current sources in addition to photoemission, however, must pay particular attention to its geometry, orientation, and material properties, as discussed in Chapter 1.1.2, because insulating surfaces cannot distribute charge evenly and are more prone to differential charg- ing. An experimental investigation was also performed in an attempt to determine the poten- tial threshold for electrostatic discharge to occur between JSC-1A and aluminum. After creating an electric field on the order of 10 8 V/m by biasing the aluminum plate to -4000 V and charging a 0.025 mm thick layer of JSC-1A positively by UV illumination, no elec- trostatic discharge events were detected. These results demonstrate that a larger electric field is required to generate breakdown between JSC-1A and aluminum. Thus, dusty surface charging in low-energy plasma, such as that presented in this dissertation, is not prone to electrostatic discharge. Risks of electrostatic discharge for dusty surfaces in regions of high energy particles and radiation, however, must be studied. 8.1.4 Multibody-Plasma Interactions Chapter7presentedastudyonmultibody-plasmainteractions. Thestudyfoundthatgreater differential charging occurs on the surface of an insulating surface opposite a dusty surface rather than opposite a clean conducting surface. Due to the dust layer’s dielectric properties, the dust layer cannot distribute its charge uniformly, and its differential charging enhances the differential charging on the astronaut’s arm. More importantly, the electric field strength generated between the astronaut’s arm and the surface increases further downstream regard- less of whether the surface is conducting or dusty. This indicates that there will be a length threshold on an insulating object at which the electric field will be strong enough to gen- erate arcing or breakdown. Decreasing the distance between the astronaut’s arm and the 8.2 Recommended Future Research 120 surface will also significantly increase electric field strength and increase the probability for electrostatic discharge. 8.2 Recommended Future Research Based on the scope of this research, the following work is recommended. 8.2.1 Expand Embedded-Wire Method The dust layer surface potentials and potentials within the dust layer presented in this dis- sertation were measured by embedded wires. To measure dust potentials more accurately, dust grain-sized conducting spheres electrically connected to a measurement plate outside the vacuum chamber should be used. For insulating materials, such as the Gore-Tex stud- ied in this dissertation, thin strands of conducting wire can be used. The embedded-wire method can be used with different types of “probes” depending on what the material and the experimental setup are. 8.2.2 Deep Dielectric Charging Given that the embedded-wire method can measure potentials within the dust layer and of insulating materials, the deep dielectric charging of a dusty surface due to electron beam gun irradiation would provide knowledge on the build up of charge within dust layers. This can provide insightful data on ESD events due to deep dielectric charging and assist in developing charge mitigation techniques. 8.2.3 Ion-Induced Secondary Electron Emission As opposed to the low-energy ions generated by the electrostatic gridded ion thruster in this dissertation, a proton gun or hydrogen source could simulate the solar wind plasma flow more accurately and generate ions with enough energy to produce secondary electrons. This 8.2 Recommended Future Research 121 would allow for the study of ion-induced secondary electron emission on dust grains and dusty surfaces. 8.2.4 Dusty-Surface Charging in Plasma Wake Region The electron temperature in lunar wake regions can be on the order of hundreds of eV’s. Given that permanently shadowed regions are of interest due to the presence of water ice and other volatiles, the charging of dusty conducting and insulating surfaces should be studied in a plasma wake region. 8.2.5 Charged Dust Motion Given a strong enough electric field, studying and tracking the motion of charged dust grains in plasma is of great interest. Though experiments performed in a vacuum chamber have to offset the force of gravity, it is important to understand how charged dust grains move across a plasma field. BIBLIOGRAPHY 122 Bibliography [1] S. Lai and M. Tautz, “Aspects of spacecraft charging in sunlight,” IEEE Transactions on Plasma Science, vol. 34, no. 5, p. 2053, 2006. [2] S. Lai, “A critical overview on spacecraft charging mitigation methods,” IEEE Trans- actions on Plasma Science, vol. 31, no. 6, p. 1118, 2003. [3] T. J. Stubbs, J. S. Halekas, W. M. Farrell, and R. R. Vondrak, “Lunar surface charging: A global perspective using lunar prospector data,” in in Dust in Planetary Systems, ESA SP-643, Citeseer, 2007. [4] W. Farrell, T. Stubbs, R. Vondrak, G. Delory, and J. Halekas, “Complex electric fields near the lunar terminator: the near-surface wake and accelerated dust,” Geophysical Research Letters, vol. 34, no. 14, p. L14201, 2007. [5] R. Manka, “Plasma and potential at the lunar surface,” in Photon and particle inter- actions with surfaces in space (R. Grard, ed.), Springer Netherlands, 1973. [6] J. Halekas, Y. Saito, G. Delory, and W. Farrell, “New views of the lunar plasma environment,” Planetary and Space Science, vol. 59, no. 14, p. 1681, 2011. [7] E. Grun, M. Horanyi, and Z. Sternovsky, “The lunar dust environment,” Planetary and Space Science, vol. 59, no. 14, p. 1672, 2011. [8] J. Colwell, S. Batiste, M. Horányi, S. Robertson, and S. Sture, “Lunar surface: Dust dynamics and regolith mechanics,” Reviews of Geophysics, vol. 45, no. 2, 2007. [9] T. J. Stubbs, R. R. Vondrak, and W. M. Farrell, “A dynamic fountain model for dust in the lunar exosphere,” Dust in Planetary Systems, vol. 643, pp. 185–189, 2007. [10] T. Stubbs, R. Vondrak, and W. Farrell, “Impact of dust on lunar exploration,” in Dust in Planetary Systems, 2005. [11] H. Miyamoto, H. Yano, D. Scheeres, S. Abe, O. Barnouin-Jha, A. Cheng, H. Demura, R.Gaskell, N.Hirata, M.Ishiguro, T.Michikami, A.Nakamura, R.Nakamura, J.Saito, and S. Sasaki, “Regolith migration and sorting on asteroid itokawa,” Science, vol. 316, no. 5827, p. 1011, 2007. [12] A. Barkan, N. D’Angelo, and R. Merlino, “Charging of dust grains in a plasma,” Physical Review Letters, vol. 73, no. 23, p. 3093, 1994. [13] C. Cui and J. Goree, “Fluctuations of the charge on a dust grain in a plasma,” IEEE Transactions on Plasma Science, vol. 22, no. 2, p. 151, 1994. BIBLIOGRAPHY 123 [14] M. Horanyi, S. Robertson, and B. Walch, “Electrostatic charging properties of simu- lated lunar dust,” Geophysical Research Letters, vol. 22, no. 16, p. 2079, 1995. [15] B. Walch, M. Horanyi, and S. Robertson, “Electrostatic charging of lunar dust,” in Physics of Dusty Plasmas: Seventh Workshop, 1998. [16] J. Goree, “Ion trapping by a charged dust grain in a plasma,” Physical Review Letters, vol. 69, no. 2, p. 277, 1992. [17] V. Chow, D. Mendis, and M. Rosenberg, “Role of grain size and particle velocity distribution in secondary electron emission in space plasmas,” Journal of Geophysical Research, vol. 98, no. A11, p. 19065, 1993. [18] M. Lampe, G. Joyce, G. Ganguli, and V. Gavrishchaka, “Interactions between dust grains in a dusty plasma,” Physics of Plasma, vol. 7, no. 10, p. 3851, 2000. [19] X. Wang, J. Colwell, M. Horanyi, and S. Robertson, “Charge of dust on surfaces in plasma,” IEEE Transactions on Plasma Science, vol. 35, no. 2, p. 271, 2007. [20] X. Wang, M. Horanyi, and S. Robertson, “Dust transport on a surface in plasma,” IEEE Transactions on Plasma Science, vol. 39, no. 11, p. 2730, 2011. [21] J. Meyer, R. Merlino, J. Heinrich, and S. Kim, “Flow of dusty plasma around an obstacle,” IEEE Transactions on Plasma Science, vol. 42, no. 10, p. 2690, 2014. [22] S. Hess, P. Sarrailh, J. Mateo-Velez, A. Champlain, J. Forest, B. Jeanty-Ruard, B. Thiebault, F. Honary, S. Marple, F. Cipriani, and A. Hilgers, “Spis-dust - mod- eling the interactions between spacecraft, plasma, and dust,” in AIAA SPACE 2015 Conference and Exposition, 2015. [23] J. Polansky, Laboratory investigations of the near surface plasma field and charging at the lunar terminator. PhD thesis, University of Southern California, 2013. [24] D. Vaniman, R. Reedy, G. Heiken, G. Olhoeft, and W. Mendell, “The lunar environ- ment,” in Lunar sourcebook - a user’s guide to the Moon (G. Heiken, D. Vaniman, and B. French, eds.), Cambridge University Press, 1991. [25] J. Halekas, G. Delory, R. Lin, T. Stubbs, and W. Farrell, “Lunar prospector obser- vations of the electrostatic potential of the lunar surface and its response to incident currents,” Journal of Geophysical Research, vol. 113, no. A9, p. A09102, 2008. [26] S. Lai, Fundamentals of Spacecraft Charging: Spacecraft Interactions with Space Plas- mas. Princeton University Press, 2012. [27] D. Ferguson, “New frontiers in spacecraft charging,” IEEE Transactions on Plasma Science, vol. 40, no. 2, p. 139, 2012. [28] P. Robinson and P. Coakley, “Spacecraft charging: progress in the study of dielectrics and plasmas,” IEEE Transactions on Electrical Insulation, vol. 27, no. 5, p. 944, 1992. [29] H. Garrett and A. Whittlesey, “Spacecraft charging, an update,” IEEE Transactions on Plasma Science, vol. 28, no. 6, p. 2017, 2000. BIBLIOGRAPHY 124 [30] H. Garrett, “The charging of spacecraft surfaces,” Review of Geophysics, vol. 19, no. 4, p. 577, 1981. [31] E. Whipple, “Potentials of surfaces in space,” Reports on Progress in Physics, vol. 44, no. 11, p. 1197, 1981. [32] T. Paulmier, B. Dirassen, D. Payan, and M. Van Eesbeek, “Material charging in spaceenvironment: experimentaltestsimulationandinducedconductivemechanisms,” IEEE Transactions on Dielectrics and Electrical Insulation, vol. 16, no. 3, p. 682, 2009. [33] M. Zimmerman, W. Farrell, T. Stubbs, J. Halekas, and T. Jackson, “Solar wind access to lunar polar craters: feedback between surface charging and plasma expansion,” Geophysical Research Letters, vol. 38, no. 19, p. L19202, 2011. [34] H. Hills, J. Meister, R. Vondrak, and J. Freeman, “Suprathermal ion detector experi- ment lunar-ionosphere detector,” in Apollo 15 - Preliminary Science Report, 1972. [35] B. O’Brien and D. reasoner, “Charged-particle lunar environment experiment,” in Apollo 14 - Preliminary Science Report, 1971. [36] D. McKay, G. Heiken, A. Basu, G. Blanford, S. Simon, R. Reedy, B. French, and J. Papike, “The lunar regolith,” in Lunar sourcebook - a user’s guide to the Moon (G. Heiken, D. Vaniman, and B. French, eds.), Cambridge University Press, 1991. [37] M. Horanyi, Z. Sternovsky, M. Lankton, C. Dumont, S. Gagnard, D. Gathright, E. Grun, D. Hansen, D. James, S. Kempf, B. Lamprecht, R. Srama, J. Szalay, and G. Wright, “The lunar dust experiment (ldex) onboard the lunar atmosphere and dust environment explorer (ladee) mission,” Space Science Review, vol. 185, no. 1, p. 93, 2014. [38] R. Pirich, J. Weir, D. Leyble, S. Chu, and M. DiGiuseppe, “Effects of the lunar environ- ment on space vehicle surfaces,” in Applications and Technology Conference (LISAT), 2010 Long Island Systems, 2010. [39] E. Asphaug, “Growth and evolution of asteroids,” Annual Review of Earth and Plane- tary Science, vol. 37, p. 413, 2009. [40] M.Horanyi,“Chargeddustdynamicsinthesolarsystem,” Annual Review of Astronomy and Astrophysics, vol. 34, p. 383, 1996. [41] D. Mendis, M. Rosenberg, and V. Chow, “Ionization equilibria in dusty plasma envi- ronments,” in Physics of Dusty Plasmas: Seventh Workshop, 1998. [42] D. Mendis, “Progress in the study of dusty plasmas,” Plasma Sources Science and Technology, vol. 11, p. A219, 2002. [43] A. Piel and A. Melzer, “Dusty plasmas - the state of understanding from an experi- mentalist’s view,” Advances in Space Research, vol. 29, no. 9, p. 1255, 2002. [44] P. Shukla, “Experiments and theory of dusty plasmas,” Dusty and Complex Plasmas: Basic and Interdisciplinary Research, p. 11, 2011. BIBLIOGRAPHY 125 [45] E. Whipple, T. Northrop, and D. Mendis, “The electrostatics of a dusty plasma,” Journal of Geophysical Research, vol. 90, no. A8, p. 7405, 1985. [46] J. Goree, “Charging of particles in a plasma,” Plasma Sources Science and Technology, vol. 3, no. 3, p. 400, 1994. [47] R. Hazelton, “Measurement of dust grain charging in a laboratory plasma,” IEEE Transactions on Plasma Science, vol. 22, no. 2, p. 91, 1994. [48] B. Walch, M. Horanyi, and S. Robertson, “Charging of dust grains in plasma with energetic electrons,” Physical Review Letters, vol. 75, no. 5, p. 838, 1995. [49] A. Zobnin, A. Nefedov, V. Sinel’shchikov, and V. Fortov, “On the charge of dust parti- cles in a low-pressure gas discharge plasma,” Journal of Experimental and Theoretical Physics, vol. 91, no. 3, p. 483, 2000. [50] M. Lampe, “Limits of validity for orbital-motion-limited theory for a small floating collector,” Journal of Plasma Physics, vol. 65, no. 3, p. 171, 2001. [51] Z.Sternovsky, S.Robertson, A.Sickafoose, J.Colwell, andM.Horanyi, “Contactcharg- ing of lunar and martian dust simulants,” Journal of Geophysical Research, vol. 107, no. E11, pp. 15–1, 2002. [52] M. Lampe, V. Gavrishchaka, G. Ganguli, and G. Joyce, “Effect of trapped ions on shielding of a charged spherical object in a plasma,” Physical Review Letters, vol. 86, no. 23, p. 5278, 2001. [53] B. Feuerbacher, M. Anderegg, B. Fitton, L. Laude, R. Willis, and R. Grard, “Photoe- mission from lunar surface fines and the lunar photoelectron sheath,” in Lunar and Planetary Science Conference Proceedings, vol. 3, p. 2655, 1972. [54] M. Abbas, P. Craven, J. Spann, E. West, J. Pratico, D. Tankosic, and C. Venturini, “Photoemission experiments for charge characteristics of individual dust grains,” Phys- ica Scripta, vol. 2002, no. T98, p. 99, 2002. [55] M. Abbas, D. Tankosic, P. Craven, R. Hoover, L. Taylor, J. Spann, and A. LeClair, “Photoelectric emission measurements on apollo 17 lunar dust grains,” in 37th Annual Lunar and Planetary Science Conference, 2006. [56] A. Champlain, J.-C. Mateo-Velez, J.-F. Roussel, S. Hess, P. Sarrailh, G. Murat, J.- P. Chardon, and A. Gajan, “Lunar dust simulant charging and transport under uv irradiation in vacuum: Experiments and numerical modeling,” Journal of Geophysical Research: Space Physics, vol. 121, no. 1, pp. 103–116, 2016. [57] L. Nouzak, I. Richterova, J. Pavlu, Z. Nemecek, and j.-v. n. p. y. p. Safrankova, Jana, “Investigations of photoemission from lunar dust simulant,” [58] C. Venturini, J. Span, and R. Comfort, “Preliminary results from a laboratory study of charging mechanisms in a dusty plasma,” in Physics of Dusty Plasmas: Seventh Workshop, 1998. BIBLIOGRAPHY 126 [59] D. Murphy and Y. Chiu, “Dusty plasmas in the vicinity of a large dielectric object in space,” Journal of Geophysical Research, vol. 96, no. A7, p. 11291, 1991. [60] X. Wang, M. Horanyi, and S. Robertson, “Dust transport near electron beam impact and shadow boundaries,” Planetary and Space Science, vol. 59, no. 14, p. 1791, 2011. [61] T. Sheridan, “Charging time for dust grain on surface exposed to plasma,” Physics of Plasma, vol. 113, no. 14, p. 143304, 2013. [62] J. Wang, X. He, and Y. Cao, “Modeling electrostatic levitation of dusts on the lunar surface,” IEEE Transactions on Plasma Science, vol. 36, no. 5, pp. 2459–2466, 2008. [63] H. Hsu, M. Horanyi, and S. Kempf, “Dust and spacecraft charging in saturn’s e-ring,” Earth, Planets, and Space, vol. 65, no. 3, p. 149, 2013. [64] H.-W.Hsu, M.Horányi, S.Kempf, andE.Gruen, “Spacecraftchargingnearenceladus,” Geophysical Research Letters, vol. 39, no. 6, 2012. [65] S. Hosoda, T. Okumura, J.-H. Kim, K. Toyoda, and M. Cho, “Development of 400 v solar array technology for low earth orbit plasma environment,” IEEE Transactions on Plasma Science, vol. 34, no. 5, pp. 1986–1996, 2006. [66] M. Cho, N. Miyata, and M. Hikita, “Effects of arcing on insulator surface potential in plasma: Image observation,” Journal of Spacecraft and Rockets, vol. 37, no. 1, pp. 150– 152, 2000. [67] M.Cho,R.Ramasamy,M.Hikita,K.Tanaka,andS.Sasaki,“Plasmaresponsetoarcing inionosphericplasmaenvironment: Laboratoryexperiment,” Journal of Spacecraft and Rockets, vol. 39, no. 3, pp. 392–399, 2002. [68] K. Toyoda, T. Matsumoto, M. Cho, Y. Nozaki, and M. Takahashi, “Power reduction of solar arrays by arcing under simulated geosychronous orbit environment,” Journal of Spacecraft and Rockets, vol. 41, no. 5, pp. 854–861, 2004. [69] T. Kawasaki, S. Hosoda, J. Kim, M. Cho, åůİåťŐåťĞ, and çťřçŤřèĄąåŔš, “Esd test of large solar array coupon in geo plasma environment,” in 9th Spacecraft Charging Technology Conference 9th Spacecraft Charging Technology Conference, p. 95, 2005. [70] T. Kawasaki, S. Hosoda, J. Kim, K. Toyoda, and M. Cho, “Charge neutralization via arcing on a large solar array in the geo plasma environment,” IEEE Transactions on Plasma Science, vol. 34, no. 5, pp. 1979–1985, 2006. [71] M. Cho, H. Masui, S. Iwai, T. Yoke, and K. Toyoda, “Three hundred fifty volt photo- voltaic power generation in low earth orbit,” Journal of Spacecraft and Rockets, vol. 51, no. 1, pp. 379–381, 2013. [72] K. Toyoda, A. Kawano, S. Miyazaki, and M. Cho, “Flashover discharge measurement with uniform surface charging and modeling of current waveform,” IEEE Transactions on Plasma Science, vol. 43, no. 9, pp. 3064–3069, 2015. [73] I. Langmuir, “Volume 4,” in The collected works of Irving Langmuir (C. Suits, ed.), Pergamon Press, Inc., 1961. BIBLIOGRAPHY 127 [74] D. Bohm, “Chapter 3,” in Characteristics of electrical discharges in magnetic fields (A. Guthrie and W. R., eds.), McGraw-Hill Book Company, Inc., 1949. [75] J.Allen,R.Boyd,andP.Reynolds,“Thecollectionofpositiveionsbyaprobeimmersed in a plasma,” Proceedings of the Physical Society, vol. B70, no. 3, p. 297, 1956. [76] S. Lam, “Unified theory for the langmuir probe in a collisionless plasma,” The Physics of Fluids, vol. 8, no. 1, p. 73, 1965. [77] F. Chen, “Langmuir probe analysis for high density plasmas,” Physics of Plasmas, vol. 8, no. 6, p. 3029, 2001. [78] R. Merlino, “Understanding langmuir probe current-voltage characteristics,” American Journal of Physics, vol. 75, no. 12, p. 1078, 2007. [79] J. Allen, “Probe theory - the orbital motion approach,” Physica Scripta, vol. 45, no. 5, p. 497, 1992. [80] R. Estes and J. Sanmartin, “Cylindrical langmuir probes beyond the orbital-motion- limited regime,” Physics of Plasmas, vol. 7, no. 10, p. 4320, 2000. [81] J. Smith, N. Hershkowitz, and P. Coakley, “Inflection-point method of interpret- ing emissive probe characteristics,” Review of Scientific Instruments, vol. 50, no. 2, pp. 210–218, 1979. [82] C. LaMarche, J. Curtis, and P. Metzger, “Permeability of jsc-1a: a lunar soil simulant,” Icarus, vol. 212, no. 1, p. 383, 2011. [83] O. Calla and I. Rathore, “Study of complex dielectric properties of lunar simulants and comparison with apollo samples at microwave frequencies,” Advances in Space Research, vol. 50, no. 12, p. 1607, 2012. [84] P. Ehrlich, L. Amborski, and R. Burton, “Dielectric properties of teflon from room temperature to 314 and from frequencies of 10 2 to 10 5 c/s,” Journal of Research of the National Bureau of Standards, vol. 51, no. 4, pp. 185–188, 1953. [85] X. Wang, M. Horányi, and S. Robertson, “Dust transport near electron beam impact and shadow boundaries,” Planetary and Space Science, vol. 59, no. 14, pp. 1791–1794, 2011. [86] M. Cho, J.-H. Kim, S. Hosoda, Y. Nozaki, T. Miura, and T. Iwata, “Electrostatic discharge ground test of a polar orbit satellite solar panel,” IEEE Transactions on Plasma Science, vol. 34, no. 5, pp. 2011–2030, 2006. [87] A. R. Khan, T. Sumida, M. Iwata, K. Toyoda, M. Cho, and T. Fujita, “Environ- ment exposure tests of electron-emitting film for spacecraft charging mitigation,” IEEE Transactions on Plasma Science, vol. 40, no. 2, pp. 380–387, 2012. [88] H. Masui, K. Toyoda, and M. Cho, “Electrostatic discharge plasma propagation speed on solar panel in simulated geosynchronous environment,” IEEE Transactions on Plasma Science, vol. 36, no. 5, pp. 2387–2394, 2008. BIBLIOGRAPHY 128 [89] A. Takahashi, R. Muraguchi, M. Iwata, and M. Cho, “Charging and arcing test on semiconductive coated solar coupon panel,” IEEE Transactions on Plasma Science, vol. 42, no. 2, pp. 384–390, 2014. [90] N. W. Green, A. R. Frederickson, and J. Dennison, “Experimentally derived resistivity for dielectric samples from the crres internal discharge monitor,” IEEE transactions on plasma science, vol. 34, no. 5, pp. 1973–1978, 2006. [91] N. W. Green and S. F. Dawson, “Electrostatic discharge testing of carbon composite solararraypanelsforuseinthejovianenvironment,” in AIAA SPACE 2015 Conference and Exposition, p. 4558, 2015. [92] P. Swaminathan, J. Dennison, A. Sim, J. Brunson, E. Crapo, and A. Frederickson, “Comparison of classical and charge storage methods for determining conductivity of thin film insulators,” 2004. [93] W. Yu, J. Wang, and K. Chou, “Laboratory measurement of lunar regolith simulant surface charging in a localized plasma wake,” IEEE Transactions on Plasma Science, vol. 43, no. 12, pp. 4175–4181, 2015. [94] D. Han, Particle-in-cell Simulations of Plasma Interactions with Asteroidal and Lunar Surfaces. PhD thesis, University of Southern California, 2015. [95] D. Han, J. Wang, and X. He, “A nonhomogeneous immersed-finite-element particle-in- cell method for modeling dielectric surface charging in plasmas,” IEEE Transactions on Plasma Science, vol. 44, no. 8, pp. 1326–1332, 2016. [96] J. Wang and D. Hastings, “Ionospheric plasma flow over large high voltage space plat- forms. ii: The formation and structure of plasma wake,” Physics of Fluids B: Plasma Physics, vol. 4, pp. 1615–1629, 1992. [97] A. Dove, M. Horanyi, X. Wang, M. Piquette, A. R. Poppe, and S. Robertson, “Experi- mental study of a photoelectron sheath,” Physics of Plasmas, vol. 19, no. 4, p. 043502, 2012. [98] J. A. Samson and R. Cairns, “Photoelectric yield of aluminum from 300 to 1300 å,” Review of Scientific Instruments, vol. 36, no. 1, pp. 19–21, 1965. [99] M. Abbas, D. Tankosic, P. Craven, J. Spann, A. LeClair, and E. West, “Lunar dust charging by photoelectric emissions,” Planetary and Space Science, vol. 55, no. 7, pp. 953–965, 2007. [100] R.E.Jorgenson, L.K.Warne, A.A.Neuber, J.Krile, J.Dickens, andH.G.Krompholz, “Effect of dielectric photoemission on surface breakdown: An ldrd report,” Sandia Report SAND2000-3044, 2003. [101] L. Cranberg, “The initiation of electrical breakdown in vacuum,” Journal of Applied Physics, vol. 23, no. 5, pp. 518–522, 1952. [102] J.Wang,P.Leung,H.Garrett,andG.Murphy,“Multibody-plasmainteractions: charg- ing in the wake,” Journal of Spacecraft and Rockets, vol. 31, no. 5, pp. 889–894, 1994.

Abstract (if available)

Abstract The objective of this dissertation is to study the effects of dust coverage on in plasma through experimental, analytical, and numerical investigations. Specifically, this dissertation investigates: ❧ 1. The effects of dust coverage on plasma charging of both conducting and insulating surfaces under plasma conditions similar to that at the lunar terminator. ❧ 2. The effects of dust coverage on photoelectron charging of both conducting and insulating surfaces, similar to that on the lunar dayside. ❧ 3. The effects of dust coverage on plasma charging under multibody-plasma interaction conditions. ❧ To study plasma charging at the lunar terminator, an electrostatic gridded ion thruster was used to generate a mesothermal plasma, which impinged the target surface at a zero degree angle of attack. Due to the lack of viable methods to directly measure dielectric surface potentials in plasma, a novel dielectric surface potential measurement technique was first developed. Both the effects of dust quantity and of the dust coverage pattern on the charging of a surface in a mesothermal plasma were analyzed. It was found that the dust layer acts as a capacitor and drives the potential of the surface beneath it more negatively. Additionally, patches of dust accumulation create plasma wake regions behind the dust patches. For a conducting surface, these plasma wake regions increase electron current collection, which drives the potential of the conducting plate further negative. An insulating surface, on the other hand, cannot distribute charge evenly and develops greater differential charging on its surface due to the combination of the dust layer capacitance and the plasma wake effects. It was found that the pattern of dust coverage on a surface affected the charging more than the quantity of dust coverage. I also validated the results with numerical modeling. ❧ To study photoelectron charging on the lunar dayside, two UV-C germicidal bulbs were used to generate ultraviolet radiation, which generated photoelectron current from the target surface. The results showed that as the dust coverage increased on a conducting plate, both the conducting plate and the dust-surface charged less positively. The same was true for an insulating material covered by dust. Additionally, the potential difference between the dust surface and the conducting or insulating surface was approximately the difference between the work functions of the two materials. Because the difference in the work functions between such materials was only a few eV's, my results showed that differential charging was not a concern for a surface that was charged only by UV-illumination. Dusty surfaces exposed to other current sources in addition to photoemission, however, might exhibit more significant differential charging, and their geometry, orientation, and material properties must be considered. The potential threshold for electrostatic dischrage (ESD) to occur between a layer of lunar dust simulant, JSC-1A, and aluminum was also studied. ❧ To study dusty surface charging under the multibody-plasma interaction conditions, an experimental setup placing an ""astronaut's arm""' in close proximity to both a conducting and a dusty surface in a mesothermal plasma was considered. This scenario is representative of an astronaut reaching over a surface to operate a piece of sensitive equipment or to pick up a tool. It was found that significantly greater differential charging occurred on an astronaut's arm when it was in close proximity to the dusty surface. Since a dusty layer does not redistribute the charge stored to achieve a uniform surface potential, it increased the differential charging on the astronaut's arm. More importantly, it was found that the electric field strength between the astronaut arm and both the conducting and insulating surfaces increased further downstream of the arm. Depending on the length of the astronaut's arm and the distance between the arm and the surface, electrical breakdown or arcing could become a significant concern.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Experimental and numerical investigations of charging interactions of a dusty surface in space plasma

PDF

Laboratory investigations of the near surface plasma field and charging at the lunar terminator

PDF

Numerical and experimental investigations of dust-plasma-asteroid interactions

PDF

The space environment near the Sun and its effects on Parker Solar Probe spacecraft and FIELDS instrumentation

PDF

Kinetic studies of collisionless mesothermal plasma flow dynamics

PDF

Grid-based Vlasov method for kinetic plasma simulations

PDF

Interactions of planetary surfaces with space environments and their effects on volatile formation and transport: atomic scale simulations

PDF

Systems engineering and mission design of a lunar South Pole rover mission: a novel approach to the multidisciplinary design problem within a spacecraft systems engineering paradigm

PDF

Particle-in-cell simulations of kinetic-scale turbulence in the solar wind

PDF

An experimental investigation of cathode erosion in high current magnetoplasmadynamic arc discharges

PDF

The cathode plasma simulation

PDF

Investigation of thin film aluminum filter contamination by monomethylhydrazine (MMH)–nitrogen tetroxide (NTO) plumes

PDF

Three-dimensional kinetic simulations of whistler turbulence in solar wind on parallel supercomputers

PDF

Numerical and experimental investigations of ionic electrospray thruster plume

PDF

An investigation into magnetically induced extractor-less electrospray propulsion devices

PDF

A study of ignition effects on thruster performance of a multi-electrode capillary discharge using visible emission spectroscopy diagnostics

PDF

Revised calibration of a long-term solar extreme ultraviolet irradiance data set

PDF

Trajectory mission design and navigation for a space weather forecast

PDF

Plasma emission across metal nanoparticle surfaces and semiconductor -liquid interface characterizations

PDF

Techniques for analysis and design of temporary capture and resonant motion in astrodynamics

Asset Metadata

Creator Chou, Kevin Lin (author)

Core Title Experimental investigation on dusty surface charging in plasma

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Astronautical Engineering

Publication Date 10/19/2017

Defense Date 10/13/2017

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag charging diagnostics,dust charging,dusty plasma,OAI-PMH Harvest,spacecraft charging

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Wang, Joseph (committee chair), Gruntman, Mike (committee member), Rhodes, Edward (committee member)

Creator Email choukl@usc.edu,kevin.lin.chou@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-445916

Unique identifier UC11263821

Identifier etd-ChouKevinL-5848.pdf (filename),usctheses-c40-445916 (legacy record id)

Legacy Identifier etd-ChouKevinL-5848.pdf

Dmrecord 445916

Document Type Dissertation

Rights Chou, Kevin Lin

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA