University of Southern California Dissertations and Theses

Multiphoton laser-induced fluorescence for measuring point specific densities of ground state atomic hydrogen in an arcjet plume

(USC Thesis Other)

Multiphoton laser-induced fluorescence for measuring point specific densities of ground state atomic hydrogen in an arcjet plume

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS This manuscript has been reproduced from the microfihn master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs included in the original manuscript have been reproduced xerographically in this copy. EBgher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Infonnation Company 300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. NOTE TO USERS The original document received by UMI contains pages with slanted print. Pages were microfilmed as received. This reproduction is the best copy available UMI R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. M u l t ip h o t o n L a s e r In d u c e d F l u o r e s c e n c e f o r M e a s u r i n g Po in t S p e c ih c D e n s it ie s o f G r o u n d S t a t e a t o m ic H y d r o g e n in a n A r c j e t Pl u m e by Jeffrey Alan Pobst A Dissertation Presented to T h e FACULTY o f t h e G r a d u a t e S c h o o l U n i v e r s i t y o f S o u t h e r n C a l i f o r n i a In Partial Fulfillment of the Requirements for the Degree D o c t o r o f P h il o s o p h y (Aerospace Engineering) December 1997 Copyright 1997 Jeffrey Alan Pobst R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. UMI Number: 9835117 UMI Microform 9835117 Copyright 1998, by UMI Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. UMI 300 North Zeeb Road Ann Arbor, MI 48103 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES. CALIFORNIA 90007 This dissertation, written by .............................................. under the direction of k.Xs Dissertation Committee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillment of re quirem ents for the degree of D O C TO R OF PHILOSOPHY Date . . 5 i;ïW R p .ç .r ..Z^..TJ7.7.T.... DISSERTATION COMMITTEE Chairperson R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Acknowledgments There are many people who deserve great thanks for their efforts in helping me with not only this particular project, but my work and life in general. I thank Ron Spores for bringing me out into the desert (honest, really) where I found that opportunity exists in the midst of hot sun, burning sand, and military procurement and safety procedures. His encouragement and support created a working atmosphere that is both professional and one of friendship. Ingrid Wysong deserves special thanks as her desire for collaboration and cross-discipline research served as the impetus for this entire project. The value of her advisory role and parmership in this entire process exceeds measure and I know that I have drawn upon her extra effort when things continued to look far from promising. Somehow this project kept going even in the bleakest of times, and those times, unfortunately, were many. Dan Erwin deserves great thanks for being an excellent advisor. I can not imagine working as a graduate student for anyone else and getting this far. He makes research more interesting and more intriguing to me than anyone I have ever met. He teaches both as a professor and as a partner. Professors Muntz and Kune have also been instrumental in both my education and my approach to research. Thank you both for the time you have spent with me and the things that I have picked up from you thus far. There are many more people whom I thank including my family (Mom, Amy, Craig, Deke, Mere), those at Phillips Lab (Daron, Dennis, Keith, Alan, Daves: White, Campbell, & Weaver; Greg, Jaime and many more), people at USC (Elsie, Alice, Gerald, John, Radek, Mike; Profs. Egolfopoulos, Cheng, Miller, Wilcox, so many more), my close friends (in the U.S., England, the Gambia, and elsewhere), the Penoliars, and everyone who has given their support for the past several years. Thanks too, to the Air Force Office of Scientific Research and Dr. Mitat Birkan who supported this work both at USC and at the Phillips Laboratory. SCENE 1: (INT NIGHT: laboratory) A graduate student is standing behind a laser wondering why light refuses to come out the front, a dismayed groan is heard as a bright red liquid pours out onto the floor from the insides o f the laser. Cut to... R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table of Contents Acknowledgments___________________________________________________________ii List of Figures ____________________________________________________________ vii Abstract__________________________________________________________________ xii 1 Introduction _______________________________________________________________ 1 2 Approach__________________________________________________________________ 4 2.1 Laser Induced Fluorescence Methods for Detecting Hydrogen Atoms ________________4 2.1.1 Laser Induced Fluorescence_______________________________________________4 2.1.2 Single Photon Laser Induced Fluorescence___________________________________ 7 2.1.3 Two Photon Excitation of Hydrogen________________________________________10 2.1.4 Three Photon Excitation of Hydrogen_______________________________________13 2.2 Detecting Hydrogen Atoms in an Arcjet Plume__________________________________ 16 2.2.1 General Issues__________________________________________________________ 16 2.2.1.1 Beam Alignment____________________________________________________ 17 2.2.1.2 Laser Power________________________________________________________ 18 2.2.1.3 Detection of Fluorescence ____________________________________________ 18 2.2.1.4 Resonant Scatter____________________________________________________ 19 2.2.1.5 Measuring the Collection Volume_____________________________________ 20 2.2.1.6 Positioning in the Plume______________________________________________21 2.2.1.7 Modeling the Raw D ata______________________________________________ 21 2.2.2 Density Measurement Issues______________________________________________ 22 2.2.2.1 Broadening Mechanisms _____________________________________________22 2.2.2.2 Collisional Quenching _______________________________________________ 24 2.2.2.3 Saturation, Multiphoton lonizauon, and Amplified Spontaneous Emission 25 2.2.3 Temperature Measurement Issues__________________________________________25 2.2.3.1 Doppler Broadening_________________________________________________ 25 2.2.3.2 Stark Broadening___________________________________________________ 26 2.2.4 Velocity Measurement Issues_____________________________________________27 2.2.4.1 Doppler Shift_______________________________________________________27 2.3 Calibration for Absolute Density Measurements_________________________________ 28 2.3.1 Atomic Hydrogen Concentrations in a Calibration C e ll_______________________ 28 2.3.1.1 Chemical Reactions in the Cell________________________________________ 32 2.3.2 Comparing Calibration Cell and Arcjet Density Measurements_________________ 33 3 Experimental Setup________________________________________________________ 34 m R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 3.1 Overview_________________________________________________ 34 3.2 Laser Equipment________________________________________________________ 36 3.2.1 Nd:YAG Pulsed L a se r___________________________________________________37 3.2.2 Tunable Dye L aser______________________________________________________ 37 3.2.3 Wavelength Doubling & Tripling______________________________________ 38 3.2.4 Wavelength Separation___________________________________________________39 3.2.5 Energy Attenuation_____________________________________ 39 3.2.6 Measured Laser Li ne width of Final Beam ___________________________________40 3.3 Beam Power/Energy Measurement____________________________________________ 40 3.4 Microwave Discharge Cavity for Calibration Cells_______________________________ 41 3.5 Gated Photomultiplier Detection______________________________________________ 42 3.6 Arcjet System ______________________________________________________________43 3.7 Three-Axis Spatial Arcjet Positioning System____________________________________45 3.8 Data Acquisition Equipment_________________________________________________ 46 4 Results and Analysis___________________________________________________________ 49 4.1 Data Verification______________________________________________ 49 4.1.1 Detection of Fluorescence at Varying Laser W avelengths______________________49 4.1.2 Detection of Fluorescence at Varying Laser P ow er___________________________ 52 4.1.2.1 Losses in Fluorescence Due to Amplified Spontaneous Em ission____________54 4.1.2.2 Temperature Prediction Effects When Fluorescence is Not Proportional to Laser Power Squared____________________________________________ 55 4.1.3 Lifetime Measurements to Determine the Effects of Collisional Quenching_______ 57 4.1.4 Determining Atomic Hydrogen Concentration in the Calibration Cell____________ 61 4.2 Ground State Atomic Hydrogen Measurements at 1.34 kW Operation_______________ 63 4.2.1 Overview _________________________________________________________63 4.2.2 Near Nozzle Exit________________________________________________________66 4.2.2.1 Axial Velocity ______________________________________________________ 67 4.2.2.2 Radial Velocity______________________________________________________ 68 4.2.2.3 Azimuthal Velocity___________________________________________________ 70 4.12.4 Temperature_______________________________________ 71 4.2.15 Density____________________________________________________________72 4.2.3 In the Arcjet Plume 10 mm From Nozzle E x it________________________________78 4.2.3.1 Axial Velocity ______________________________________________________ 78 4.2.3.2 Radial Velocity______________________________________ 79 4.13.3 Temperature ______________________________________________________ 80 4.2.3.4 Density____________________________________________ 81 4.2.4 Along the Centerline Axis From Nozzle Exit to 30 mm Downstream_____________ 82 4.2.4.1 Axial Velocity _____________________________ 82 4.14.2 Temperature________________________________________________________ 83 4.2.4.3 Density__________________________________ 84 4.2.5 In the Nozzle Edge Wake Region__________________________________________ 85 4.2.5.1 Density_____________________________________________________________ 85 4.3 Ground State Atomic Hydrogen Measurements at 800 W O peration_________________88 iv R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.3.1 Overview______________________________________________________________ 88 4.3.2 Near Nozzle Exit________________________________________________________ 89 4.3.2.1 Axial Velocity _____________________________________________________ 90 4.3.2.2 Temperature________________________________________________________91 4.3.2.3 Density____________________________________________________________ 92 4.3.3 In the Arcjet Plume 10 mm From Nozzle E x it_______________________________ 93 4.3.3.1 Axial Velocity _____________________________________________________ 93 4.3.3.2 Temperature________________________________________________________94 4.3.3.3 Density____________________________________________________________ 95 4.3.4 Along the Centerline Axis From Nozzle Exit to 30 mm Downstream_____________ 96 4.3.4.1 Axial Velocity _____________________________________________________ 96 4.3.4.2 Temperature________________________________________________________97 4.3.4.3 Density____________________________________________________________ 98 4.4 Comparison of Measurements with Modem Computational Predictions and Other Experimental Results___________________________________________________________ 98 4.4.1 Nozzle Exit Arcjet Measurements at 1.34 kW Operation______________________ 99 4.4.1.1 Velocity Data and M odels___________________________________________ 100 4.4.1.2 Nozzle Exit Temperature vs. excited state LIP, Raman, models____________ 101 4.4.1.3 Nozzle Exit Density vs. model results_________________________________ 103 4.4.2 Arcjet Plume at Two Power Levels - Measurements and Modeling Results_______105 4.4.2.1 Overview_________________________________________________________ 105 4.4.2.2 Nozzle Exit________________________________________________________ 105 4.4.2.3 10 mm Downstream________________________________________________ 109 4.4.2.4 Axial Profile______________________________________________________ 113 4.5 Analysis of Nozzle Exit Data at 1.34 kW Operation and Resulting Implications to .Arcjet Performance Param eters______________________________________________________ 116 4.5.1 Specific Impulse from Density and Velocity Measurements___________________117 4.5.2 Conservation of Total Mass Flow_________________________________________124 4.5.3 Molecular Dissociation Fraction at the Arcjet Nozzle Exit____________________ 125 4.5.4 Energy Loss Due to Molecular Dissociation in the Hydrogen Arcjet____________ 126 5 In Conclusion _______________________________________________________________ 130 1 Literature Review Appendix________________________________________________ 132 1.1 Research on Arcjet Thrusters ______________________________________________ 132 1.1.1 NASA Lewis Research C enter___________________________________________ 133 1.1.2 Rocket Research Company / Olin Aerospace Company / Primex Aerospace Company__________________________________________________________________134 1.1.3 Air Force Phillips Laboratory / University of Southern California______________ 136 1.1.4 NASA Jet Propulsion Laboratory_________________________________________138 1.1.5 Institut Filer Raumfahnsysteme at Universitat Stuttgart (Germany)____________ 139 1.1.6 Aerospace Corporation_________________________________________________ 141 1.1.7 BPD Difesae Spazio (Italy)_____________________________________________ 142 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1.1.8 Stanford University____________________________________________________ 143 1.1.9 University of Tennessee Space Institute____________________________________145 1.1.10 Massachusetts Institute of Technology____________________________________146 1.1.11 Texas Tech University________________________________________________ 147 1.1.12 University of Illinois__________________________________________________ 148 1.1.13 Ohio State University_________________________________________________ 148 1.1.14 University of Michigan________________________________________________ 149 1.1.15 Cornell University_____________________________________________________150 1.2 Discussion on Arcjet Research ______________________________________________ 152 1.3 Laser Detection of Atomic Hydrogen_________________________________________ 154 1.3.1 Sandia National Laboratory_____________________________________________ 155 1.3.2 Institut fur Physikalische Chemie der Verbreimung (Germany)________________ 158 1.3.3 Ohio State University__________________________________________________ 159 1.3.4 Kyushu University (Japan)______________________________________________ 159 2 References Appendix ________________________________________________ 161 V I R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. List of Figures Figure 2-1: Interaction (in this case an absorption o f an incoming photon) with a molecular or atomic species can cause an excitation o f an electron in a lower electronic state to be excited to an upper state and consequently emit a photon of a frequency corresponding to the transition of relaxation back to the lower level.____________________________________________________ 5 Figure 2-2: A simplified two level energy diagram showing the excitation and relaxation processes__________________________________________________________________________ 7 Figure 2-3: Two photon excitation o f the ground state hydrogen atom to n=3 and the resulting fluorescence (not to scale).__________________________________________________________ 12 Figure 2-4: Two-step, three photon excitation schemes (not to scale). _____________________14 Figure 2-5: Density Calibration cell schematic._______________________________________ 30 Figure 3-1: Experimental setup._____________________________________________________34 Figure 3-2: Simple Arcjet Cutaway Diagram__________________________________________44 Figure 4-1: Sample 2PL1F spectrum o f the hydrogen atom, showing the calibration signal from the discharge cell and the signal from the arcjet plume along with Gaussian least square fit profiles.__________________________________________________________________________ 50 Figure 4-2. Arcjet study o f LIF signal dependence on laser power compared to ASE power dependence, which has a much sharper response to laser power.__________________________ 53 Figure 4-3. Narrower ASE spectrum compared with U F spectrum. ______________________ 54 Figure 4-4. Apparent temperature fo r the arcjet at changing laser powers_________________ 56 Figure 4-5: Fluorescence decay fo r n=3 hydrogen atoms at two positions along the arcjet nozzle exit plane. For comparison, a trace o f the laser pulse alone (with no hydrogen atoms present) is shown.__________________________________________________________________________ 57 Figure 4-6: Fluorescence lifetimes as a function of radial position along the arcjet nozzle exit ( 0.4 mm away) from many different days. The line represents a best fit to the d a ta __________59 Figure 4-7: Fluorescence lifetimes as a function o f axial distance from the arcjet nozzle exit along the plume centerline from three different days. The line represents a best fit to the data 59 Figure 4-8: Fluorescence lifetimes as a function of radial distance taken 10 mm downstream from the arcjet nozzle exit on three different days. The line represents a best fit to the d a ta 60 vu R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Figure 4-9. Determination o f atomic hydrogen density and emitted signal through titration of NO 2 into calibration cell.____________________________________________________________61 Figure 4-10: V-I Curve fo r 1 kW hydrogen arcjet operating with 13.1 mg/s o f hydrogen. For comparison, the V-l curve with the same hydrogen flow is shown fo r the Stanford apparatus. The data is based on specific energy data published by Olin Aerospace regarding the Stanford measurements. ____________________________________________________________________ 64 Figure 4-11: A visualization o f the physical location where data was taken in the arcjet plume. 65 Figure 4-12. Profile across nozzle exit plane (0.4 mm from exit) o f the œàal velocity component o f the ground state hydrogen atoms containing data from several different da ys.____________ 67 Figure 4-13. Profile across nozzle exit plane (0.4 mm from exit) o f the radial velocity component o f the groimd state hydrogen atoms containing data from several different days where the radial position is measured parallel to the incoming laser beam .________________________________69 Figure 4-14. Profile across nozzle exit plane (0.4 mm from exit) o f the azimuthal velocity component (indicating swirl) o f the ground state hydrogen atoms containing data from several different days where the radial position is measured perpendicular to the incoming laser be am. 70 Figure 4-15: Temperature profile across the nozzle exit plane (0.4 mm from exit) o f the ground state hydrogen atoms. Graph contains data from several different days. ___________________ 71 Figure 4-16. Significance in correcting fo r quenching when determining density from a relative fluorescence profile. Note that the largest correction occurs in the center o f the profile where the quenching due to collisions with the hydrogen molecules or electrons is greatest and is minimal at the edges where quenching is relatively insignificant.__________________________ 73 Figure 4-17: Profile across nozzle exit plane (0.4 mm from exit) o f the relative number density of ground state hydrogen atoms containing data from several different days.__________________74 Figure 4-18: Profile across nozzle exit plane (0.4 mm from exit) o f the absolute number density o f ground state hydrogen atoms measured on a single day. _______________________________76 Figure 4-19. Profile across nozzle exit plane (0.4 mm from exit) o f the absolute number density o f ground state hydrogen atoms containing data from several different days.________________77 Figure 4-20: Radial profile o f ground state hydrogen atom axial velocity across the arcjet plume, 10 mm downstream o f (and parallel to) the nozzle exit plane. Plot contains data from multiple days._____________________________________________________________________________ 78 Figure 4-21: Radial profile o f ground state hydrogen atom radial velocity across the arcjet plume, 10 mm downstream o f (and parallel to) the nozzle exit plane. Plot contains data from midtiple days.______________________________________________________________________79 vui R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Figure 4-22: Radial temperature profile of ground state hydrogen atoms across the arcjet plume, 10 mm downstream o f (and parallel to) the nozzle exit plane. Plot contains data from multiple days._____________________________________________________________________ 80 Figure 4-23: Radial profile o f ground state hydrogen atom density across the arcjet plume, 10 mm downstream o f (and parallel to) the nozzle exit plane. Plot contains data from midtiple days.____________________________________________________________________________ 81 Figure 4-24: Axial profile o f ground state hydrogen atom velocity along the arcjet plume centerline. Plot contains data from several different days._______________________________ 83 Figure 4-25: Axial profile o f ground state hydrogen atom temperature along the arcjet plume centerline. Plot contains data from several different days._______________________________ 84 Figure 4-26: Axial profile o f ground state hydrogen atom temperature along the arcjet plume centerline. Plot contains data from several different days._______________________________ 85 Figure 4-27: Arcjet Nozzle Wake Region Density map__________________________________ 87 Figure 4-28: Profile across the nozzle exit plane (0.4 mm from exit) o f the axial velocity component o f the ground state hydrogen atoms containing data from several different days. 90 Figure 4-29: Temperature profile across the nozzle exit plane (0.4 mm from exit) o f the groimd state hydrogen atoms. Graph contains data from several different days. __________________ 91 Figure 4-30: Profile across nozzle exit plane (0.4 mm from exit) o f the absolute number density o f groimd state hydrogen atoms containing data from several different d ays._______________ 92 Figure 4-31: Radial profile o f ground state hydrogen atom axial velocity across the arcjet plume, 10 mm downstream o f (and parallel to) the nozzle exit plane. Plot contains data from multiple days._____________________________________________________________________ 93 Figure 4-32: Radial temperature profile of ground state hydrogen atoms across the arcjet plume, 10 mm downstream o f (and parallel to) the nozzle exit plane. Plot contains data from mtdtiple days._____________________________________________________________________ 94 Figure 4-33: Radial profile o f groimd state hydrogen atom density across the arcjet plume, 10 mm downstream o f (and parallel to) the nozzle exit plane. Plot contains data from multiple days._____________________________________________________________________________95 Figure 4-34: Axial profile o f ground state hydrogen atom velocity along the arcjet plume centerline. Plot contains data from multiple days.______________________________________96 Figure 4-35: Axial profile o f ground state hydrogen atom temperature along the arcjet plume centerline. Plot contains data from several different days._______________________________ 97 IX R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Figure 4-36: Axial profile o f ground state hydrogen atom temperature along the arcjet plume centerline. Plot contains data from several different days._______________________________ 98 Figure 4-37. Profile across nozzle exit plane (0.4 mm from exit) o f the axial velocity component o f the ground state hydrogen atoms containing data from several different days. For comparison, corresponding modeling and experimental data is sho\vn.^‘' ‘'*"''^'^^''^_________ lOl Figure 4-38: Temperature data (previously shown in Figure 4-15) with corresponding data (same arcjet conditions, same profile location) fo r U F o f the electronically excited hydrogen atoms,‘'* Raman molecidar rotational temperatures,^' and Computational models. i02 Figure 4-39: Groimd state hydrogen density data (previously shown in Figure 4-19) with corresponding data (same arcjet conditions, same profile location) from Navier-Stokes and DSMC Computational models. J 0 4 Figure 4-40: Hydrogen velocity measurements and DSMC predictions at arcjet nozzle exit for 1.34 kW and 800 W . '^________ ^ ____________________________________________________ 106 Figure 4-41: Hydrogen temperature measurements and DSMC predictions at arcjet nozzle exit fo r 1.34 kW and 800 W. '' _________________________________________________________ 107 Figure 4-42: Hydrogen density measurements and DSMC predictions at arcjet nozzle exit for 1.34 kW and 800 W . " _______ '_____________________________________________________108 Figure 4-43: Hydrogen velocity measurements, predictions, and normalized predictions 10 mm from nozzle exit. _______________________________________________________________110 Figure 4-44: Hydrogen temperature measurements, predictions, and normalized predictions 10 mm from nozzle e.xit. ____________________________________________________________ 111 Figure 4-45: Hydrogen density measurements, predictions, and normalized predictions 10 mm from nozzle exit. _______________________________________________________________112 Figure 4-46: Hydrogen velocity measurements, predictions, and normalized predictions along the a.xial centerline. _____________________________________________________________ 113 Figure 4-47: Hydrogen temperature measurements, predictions, and normalized predictions along the axial centerline. _______________________________________________________ 114 Figure 4-48: Hydrogen density measurements, predictions, and normalized predictions along the axial centerline. _____________________________________________________________/15 Figure 4-49: Atomic hydrogen density data as seen in Figure 4-19 with polynomial fits for high, low, and mean density values. ______________________________________________________120 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Figure 4-50: Molecular hydrogen density data reprinted from Stanford University with the addition o f a polynomial fit and error bars, the latter are placed as described In the technical report o f the measurement^' _______________________________________________________ 120 Figure 4-51: Atomic hydrogen velocity data as seen In Figure 4-12 and polynomial fit o f mean velocity._________________________________________________________________________ 121 Figure 1-1 : Research Institutions and the arcjet research areas that each has participated ln.l51 XI R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Abstract This dissertation describes the use of multiphoton laser induced fluorescence (LIF) to ascertain previously unknown physical properties such as ground state hydrogen densities in arcjet space thruster plumes. Data from the implementation of the technique is given, demonstrating successful use of the technique and providing a greater understanding of the arcjet flow field at the nozzle exit and in the downstream plume. A brief discussion on LIF as a diagnostic is included, with a focus on areas of concern about the use of the technique in the arcjet environment. The experimental approach is described such that replication of this work or additional similar work may be conducted without unnecessary difficulty. Results from several diagnostic set-ups and arcjet operating conditions are given to provide an exhaustive database on ground state hydrogen properties in the arcjet plume such that the need for additional measurements of this kind on a low power hydrogen arcjet is significantly reduced. Comparisons with data from complimentary experiments and several computational models are provided to lend insight towards a better understanding of thruster operation with the addition of the data reported herein. Analysis is provided on the congruence of the measured data with previous expectations of low power arcjet behavior, and ranges of phenomena whose magnitudes were previously unknown, such as molecular dissociation, are given as based upon data provided by this work. A literature search on arcjet research in the international community is included as an additional supplement. XU R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1 Introduction Arcjets are electrically powered space rockets that convert electric power (usually from solar arrays) into directed thrust for spacecraft maneuvers. This is accomplished by passing a propellant fluid through a high energy arc of electricity to add energy to the propellant thermally, and then expanding the fluid through a nozzle to provide thrust. This process allows a spacecraft to use smaller quantities of fuel than if it used comparable chemical rockets, as the energy for propulsion is provided from the spacecraft’s solar arrays and not from chemical bonds stored in the fuel. Arcjets are expected to play an ever increasing role in satellite propulsion needs, primarily for stationkeeping and on-orbit maneuvering in the near term. While the technology is considered viable enough to be deployed on a Telstar IV communications satellite for stationkeeping, ‘ arcjet technology is far from maturity. In order to compete successfully with chemical propulsion systems for on-orbit missions, further improvements in arcjet propulsion systems are still required.' If needed improvements in the performance level and efficiency of arcjets are to be achieved, an increased understanding of the fundamental physical processes that govern the operation of an arcjet is essential. In addition, an ability to predict the plume behavior of a space propulsion device is necessary for prediction and amelioration of damaging plume-spacecraft interactions. Significant arcjet energy loss results from velocity profile losses due to thick internal boundary layers in the arcjet nozzle and from frozen flow losses such as molecular dissociation. To quantify profile losses, both gas velocity and density distributions must be known. In addition. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. arcjet models, which are necessary for timely and cost-effective improvements to design, must be tested by comparison with key physical parameters. At present, only limited inroads have been made into the problem of plume density measurements. For hydrogen arcjet thrusters, determination of species density has previously only been accomplished through the use of absorption spectroscopy techniques. ^ These VUV and XUV spectroscopy approaches are quite difficult to implement in practice and are limited to determination of line-of-sight averaged number densities at downstream locations in the plume where the optical depth is not high. These techniques have not been applicable in the determination of atomic density profiles at the thruster nozzle exit, which are desired in order to calibrate and verify recent advanced computational arcjet modeling results.^ '® The density of atoms relative to the density of molecules in the arcjet thruster is an indication of how much energy is lost into dissociation of the hydrogen molecules and not recovered through recombination into translational kinetic modes. To determine the molecular dissociation fraction and the significance of tfiis energy loss mecfianism, the molecular species density needs to be known as well as the atomic number density. The recent use of Raman spectroscopy" in an arcjet plume has provided the first information on molecular hydrogen densities at the arcjet nozzle plane. For measuring excited state atomic hydrogen velocity and temperature, excited state LIF^ '^ '^ '* has proven to be accurate and essentially nonintrusive. Though the excited states of hydrogen are more easily accessible to optical diagnostic techniques, most hydrogen atoms in the plume region are expected to be in the ground state. The ability to measure ground state atomic hydrogen temperature and velocity in addition to density allows for examination of differences in temperature R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. and velocity between the excited state and ground state species in the expected non-equilibrium arcjet plume environment Observing the need for the determination of atomic ground state hydrogen densities in the hydrogen arcjet plume for ongoing modeling efforts and for determining the significance of frozen flow losses such as molecular dissociation, this dissertation presents the theoretical approach for measuring the atomic ground state hydrogen atoms in the arcjet plume. An experimental description and design is included followed by experimental data taken using multiphoton laser induced fluorescence. The data is taken at several locations in the arcjet plume at varying arcjet operating conditions and the results of these experiments are provided. Analysis and conclusions drawn from this data are discussed, and the results are compared with previous experimental and modeling efforts in order to increase the understanding of the physical flow behavior of the low power hydrogen arcjet thruster. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2 Approach 2.1 L a se r In d u c e d F l u o r e sc e n c e M e t h o d s fo r D e t e c t in g H y d r o g e n A t o m s 2. /. I L a s e r in d u c e d F l u o r e sc e n c e Laser induced fluorescence (LIF) spectroscopy has gained wide acceptance in the past three decades as a diagnostic technique that generates measurable fluorescence from a small species population that otherwise might not be detectable.'^ Fluorescence, the spontaneous emission of radiation from upper energy levels, is observed when a population increase of an upper level is created through photon absorption, particle collision, or other interaction, and then emits photons as the level relaxes to a lower population (see Figure 2-1 ). Increasing the population of specific upper levels can be obtained by exciting specific lower levels of the species with coherent light of fixed frequency. This excitation process (usually accomplished with laser light) can increase the population of the upper level for a set duration based upon the population of the lower level. Fluorescence resulting from the increased population during the pumping time is often significantly larger than without the excitation pumping and thus provides a means of observing small species populations and allows investigation of the populations of the lower energy levels that supply the upper levels. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Figure 2-1: Interaction (in this case an absorption of an incoming photon) with a molecular or atomic species can cause an excitation of an electron in a lower electronic state to be excited to an upper state and consequently emit a photon of a frequency corresponding to the transition of relaxation back to the lower level. Laser light used for the excitation process must be of the specific frequency that corresponds to the energy difference between the lower and upper levels of the transition being probed. In order to create laser light at the specific frequency required for many different transitions, tunable dye lasers are often employed to allow probing in the range from 200nm to 1.5p.m. Probing in regions around and below 200nm is often difficult due to atmospheric absorption, but transitions in this regime can be induced without using photons near and below 200 nm through the use of multiphoton processes. Multiple photons of a frequency related to a specific transition spacing by an integer multiplication, can induce the transition to take place. A weaker excitation cross section is typical for multiphoton transitions when compared to the cross section for the transition with single photons. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. To perform LIF (with single or multiple photon transitions) on a particular species, several criteria must be met. The molecule or atom in question must have a known emission spectrum so that the fluorescence detected from the excitation can be properly accounted for. Processes such as dissociation from the excited state can affect the fluorescence that is measiu-ed, and sometimes prevent it from being detected at all. This is also important if the fluorescence produced from exciting the species is to be related to the intensity of the pumping or the density of the species in the lower levels. As previously discussed, methods for creating the light needed to pump the transition must also be available. Tunable coherent light sources are desired for use in LIF due to the need to both match frequencies to atomic and molecular transitions and to scan across the transition frequencies in order to leam information about fine structure, intensity, linewidth, etc. In different spectral ranges, different tuning methods have been developed to initiate the desired light. Devices spanning these ranges include semiconductor diode lasers, tunable gas lasers, pumped dye lasers (both pulsed and continuous wave), and excimer lasers.'* The radiative decay rate from the upper excited state must be known as the fluorescence power is proportional to that rate.'* Also, as other species are present, a decrease in excited state population may occur due to factors other titan spontaneous emission such as collisions and laser- induced chemistry. These quenching processes must be accounted for in order to correct the measured fluorescence to determine species ntunber density. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2.1.2 S in g l e P h o t o n L a s e r I n d u c e d F l u o r e s c e n c e n=2 n=l Figure 2-2: A simplified two level energy diagram showing the excitation and relaxation processes In the simplified case of a two level energy system with a lower level (n=l) and an upper level (n=2), the rates for the optical and collisional processes connecting the two levels (see Figure 2-2) can be described by four processes and their rate coefficients: stimulated absorption ibn), stimulated emission (£ » 2/), spontaneous emission (A), and quenctiing {Qzi). The Einstein A and B coefficients are related to these rates in the following manner. The spontaneous emission is given directly by the Einstein A coefficient. The Einstein B coefficient relates to the spontaneous emission tfirough the following: Equation 2-1 StcH n ^ A = —p — fl^(v) where h is Planck’s constant, v is the frequency of the fluorescence tfiat is emitted, c is the speed of light, and g( v) is the transition lineshape as a function of frequency. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The two stimulated rates are related to the Einstein B coefficient through the following expression: BI Equation 2-2 ft,, = ft,, = — — where h is the spectral irradiance which is defined as the incoming laser intensity per unit frequency interval and c is the speed of light. The stimulated rates are therefore a function of the incident laser intensity, and line width in addition to the nature of the species transitions in question. The rate equations for the change in population of the two states can be written utilizing these four parameters and the populations of the respective levels Ni and A f,: dN, Equation 2-3 = -yV,ft,, + jV, (ft,, + A + 0,, ) dN. Equation 2-4 = +A^,ft,, - N. (ft,, + A + g,, ) Equation 2-3 and Equation 2-4 allow the determination of each level’s population. Note that dN, dN. d f \ -T— H — :— = — (A/, + N .) = Q, so that total number density N, = N, + N. is constant. By dt d t dt^ ' ' ' - ^ substituting for N, and then integradng, a relation for the upper state population based upon an exponential term in time can be determined (see Reference 15), and is given by the following expression. Equation 2-5 A^,(r) = ---------------- A ft,, +ft,, + A + Q R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The derivation is neglected here, but note that as the exponential approaches a steady state (as the exponential term in time becomes greater than one) the population of the upper level approaches a constant value as shown in the following expression. b^^N j Equation 2-6 = lim N. (t) = This equation can be rearranged to the following using Equation 2-2. / Equation 2-7 = Nj- bxz Nr 1 + 7------— 2 b\2 where the saturation spectral intensity is defined by the following expression. A + Q Equation 2-8 / “ = c IB Thus, for the closed system case, after a time greater than the exponential characteristic time, the upper level population can be described by the ratio of spectral intensity to its saturation value. It is difficult with a more realistic and more complicated system to attain a steady state in the population of two levels and solve for the population by setting the two rates equal to zero. This is because the possibility exists that a steady state might not be achieved when, for example, laser pulses are short and laser intensities are small. Examining the fluorescence process may lead to a better understanding of the approach required for calculating the populations for differing situations. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The fluorescence power that can be observed when relaxation ttom the upper level occurs can be defined based upon the population of the upper level, the spontaneous emission rate, the firequency of the emitted photon, and the physical irradiation volume of the laser light. Equation 2-9 F = h \N . A - 4tc where /iv is the photon energy, Q is the collection solid angle, and V is the volume comprised of the focal area of the laser beam and the length over which the fluorescence is observed. For a given optical setup probing a given transition of a particular species, the fluorescence should vary then only with Ni and with the lineshape g( v) (which is dependent on frequency, temperature, electron number density, etc.). Using this expression to solve for population based upon measured fluorescence in real situations, however, neglects the effects of quenching and other depopulation processes that would cause use of the above relation to underpredict the population of N: based upon the fluorescence measurement alone. Once these other processes are taken into account. Equation 2-9 can be quite useful. 2 .1.3 T w o PHOTON E x c it a t io n o f H y d r o g e n Transitions of interest are often found where excitation would require photons with a wavelength below 200 nm. Not only is excitation of these transitions difficult due to the absence of tunable wavelength lasers in this range, but it is also difficult to implement due to the strong absorption at these wavelengths by the air in the laboratory and by other gases that may be present inside a test facility. Consequently, when transitions below 200 nm are intended to be pumped, a multiphoton approach (usually just two or three photons) is used. Two photon cross sections are quite small as the probability of two photons striking the atom or molecule at the same time and 10 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. V simultaneously providing the required energy for excitation {2h — = h v ) is substantially smaller than the probability of excitation by just one photon of the transition frequency. However, photons with half of the frequency or double the wavelength can be much easier to create in the laboratory. Since the cross section for a two photon absorption is so small, stimulated emission of an atom that has been excited by the two photon process can be usually neglected. For a two photon process. Equation 2 ^ changes to the following equation d N . E quation2-I0 = where Wa is the two photon rate given by l^C2g( v). Two photon cross sections have been measured and reported in literature'^ and cannot be directly related to single photon cross sections due to their representation of different processes and different units. As an initial and very general “rule-of-thumb." an order of magnitude more laser intensity is often required for signals similar to standard LEF when using a two photon scheme on atomic species in an atmospheric flame. Fluorescence due to a two photon absorption is quadratically related to laser power (due to the relationship on power for each photon required) and the excited state population is affected by quenching linearly as in single photon fluorescence. The specific problem of excitation of the ground state of hydrogen to its excited states requires a single photon below 200 nm or a two photon excitation using wavelengths that are greater. Since this excitation is intended for use as a diagnostic technique, detection will be important so excitation to the second excited state (n=3) or third excited state (n=4) will be desired 11 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. as relaxation to the first excited state from either of these states can be detected in the visible wavelength range (656 run for n=3 — > n=2, 486 nm for n=4 — » n=2) . Excitation to the second excited state of hydrogen and the resulting fluorescence is shown in Figure 2-3. Excitation from n=l to tfie second excited state normally requires photons of wavelength 102.5 nm to provide the correct excitation energy. Photons of wavelength at 205 nm will have a smaller chance to excite the same transition, but 205 photons can be much more easily created in the laboratory through doubling of 410 nm light or tripling of 615 nm laser light, both of which are available using tunable dye lasers. I 205 nm k \ 656 nm r n=4 n=3 n=2 205 nm n=I Figure 2-3: Two photon excitation of the ground state hydrogen atom to n=3 and the resulting fluorescence (not to scale). One of the problems of the scheme shown in Figure 2-3 is the population inversion created between the first and second excited states when directly pumping ground state atoms to the second excited state. This inversion serves as a gain medium for reabsorption of the emitted photons when relaxation occurs from n=3 to n=2 along the path of laser stimulation. This reabsorption can 12 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. amplify the spontaneous emission naturally occurring in the n=3 state and artificially depopulate n=3 at an increased rate. This mechanism (known as “amplified spontaneous emission” or ASE) can prevent the collection of fluorescence from directly relating to the number of atoms excited, making a density correlation, for example, incorrect. This ASE process can actually create a new laser beam in the hydrogen being probed along the path of the 205 nm two photon excitation beam. Examination of tfiis beam and the ASE process were investigated by researchers at Sandia National Labs.‘^ Their work is discussed in more detail in the literature overview Appendix. 2 .1.4 THREE P h o t o n E x c it a t io n o f H yd ro g en There are two methods of exciting the ground state of atomic hydrogen involving three photon excitation that are discussed in the literature overview Appendix. One of the techniques was developed at Sandia by Goldsmith’* and involves populating the first excited state with two photons, and then populating the second or third excited state with a third wavelength photon, avoiding the population inversion problem and ASE. Another three photon method involves directly populating the second excited state with three photons of the same wavelength.'^ Two step laser induced fluorescence on ground state hydrogen atoms uses two 243 nm photons to promote ground state atoms to the first excited state (normally requiring a single photon at 121.5 nm) a second photon of either 486 nm or 656 nm wavelength is used to excite the atom in the first excited state to the tfiird or second excited state respectively. This is shown in Figure lA . These two step techniques are more difficult in two respects. First, two step excitation requires resonant detectioiu In other words, the light being detected is at the same wavelength as one of the laser beams and cannot be spectrally separated. The fluorescence is only separated temporally from the laser beam on a time scale related to the Einstein coefficients of the 13 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. depopulation of the chosen excited state. Scattered laser light (which can also be shifted temporally from the initial beam after "bouncing” around the test chamber) can interfere with measurement of the fluorescence signal especially in low signal situations and make detection extremely difficult. Second, both techniques require a second photon to continue the excitation process once the first excited state is populated. This requires the two lasers to overlap the detection area both spatially and temporally in order for the full excitation to occur. In the case where the second step is accomplished by a 656 nm photon, a second laser is required. In the case using a 486 nm beam, one laser can produce 486 nm and some part of the beam can be doubled for 243 nm requiring only one laser, though it must have enough energy per pulse to support each of the transitions in the environment being probed. This method is called single laser two step excitation and is the preferred implementation of the two step techniques. 486 nm 656 nm 243 nm 243 nm - o / w w 243 nm 243 nm n=4 n=3 n=2 n=l Figure 2-4: Two-step, three photon excitation schemes (not to scale). 14 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Two step techniques avoid population inversions associated with other two and three photon excitation schemes, but add an additional layer of difficulty in both implementation and understanding of what is being probed. Instead of exciting the ground state directly to a level where fluorescence takes place, two step excitation excites atoms to an intermediate level where they are then excited again at some probability function to a higher excited state. If density measurements are to be made using this technique a method for understanding the total cross section of the excitation must be developed. Two laser lineshapes are used to promote the atom to the final excited state and a convolution of the two lineshapes are what satisfy promotion to the final excited level. This becomes even more difficult if the gas being probed is moving at high velocities (km/s) with respect to the incoming photons and Doppler shifts are significant as in the arcjet plume environment. The first beam will excite atoms based upon their relative velocity as well as transition and the second beam will also excite some part of those by velocity from a group that has already been selected by velocity. Tying the laser scan rates together by wavelength (or using one laser for both beams) may allow this problem to tie overcome. As previously mentioned there is another scheme found in the literature that involves using three photons to probe ground state hydrogen atoms. Three photons at 292 nm together excite the ground state to n=4 where n=4 to n=2 fluorescence is observed. This method has not been deemed very practical due to the even smaller cross section for three photon excitation than for two photon excitation and it does not deal with the ASE issue any better than two photon excitation to n=3. Additionally, when compared to the previously mentioned excitation schemes it is found to have lower signal to noise than the others. It does have the benefit of using the highest wavelength photons which may be more easily created in the laboratory. 15 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2.2 D e t e c t in g H y d r o g e n A t o m s in a n A r c jet Pl u m e As the point-specific density of atomic hydrogen at the nozzle exit of a 1 kW hydrogen arcjet thruster is to be undertaken using multiphoton laser induced fluorescence, an initial technique must be chosen. Of the techniques described in the prior sections, the two that appear to be of most interest are two photon direct excitation of the second excited state and single laser two step three photon excitation to the third excited state. Each of these techniques has been shown to adequately probe the flame environment for ground state atomic hydrogen and these two appear to be the best candidates for attempts in an arcjet plume to measure absolute density of hydrogen both from laboratory convenience and from comparison between techniques as described in the prior literatime section. For this work, the two photon direct excitation method is implemented, and it results in providing successful probing of the H atoms in the arcjet plume. Had ASE and other properties unique to this method been insurmountable In achieving the measurements, the other approach would have been implemented. It is shown in the following chapters that tfiis two photon direct excitation diagnostic can also determine velocities and temperatures of the ground state hydrogen in the arcjet plume using the same experimental setup as for density measurements. In order to accomplish these measurements accurately, there are several issues that must be addressed in order to acfiieve success. 2.2.1 G e n e r a l ISSUES The experiment is quite complex and many variables must be satisfied in order for data to be taken. Some general items that must be taken into account are listed first, with concerns on data interpretation following. 16 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2 .2 .1.1 B eam A lig n m e n t This point specific diagnostic technique requires that at minimum, accurate spatial alignment between one laser beam and an optical detection train be maintained. Depending upon the excitation technique, multiple pulsed beams could be required to be aligned, not only spatially, but also temporally. To make matters more difficult for alignment, at least one of the laser beams is in the ultraviolet spectrum and is not be visible. In order to satisfy this requirement a target made of a pointed metallic rod is attached to the three dimensional translation stage the arcjet is mounted on (see section 3). and set a fixed distance away from the center of the arcjet nozzle exit. When it is the time to align the laser beams, the rod is moved to the location the where detectors image, and the beams are aimed onto the rod. Ultraviolet light scattering off of the tip of the rod is visible to a photomultiplier detector without spectral filtering. Observing the signal intensity on a fast oscilloscope, temporal alignment can be made. Differentiation between laser beams can be made by blocking one or the other and observing the signal on the oscilloscope. Visually, the effect of the ultraviolet beam on the end of the metal rod can be seen as a spark with the laser equipment that has been used for the preliminary results and the ultraviolet beam can be seen to fluoresce on what is believed to be a high cotton content business card (many different cards have been tested and there is an amazing difference in susceptibility to fluorescence between them). After alignment, the motion control stages are returned to their arcjet-centric home position. Absolute position within the arcjet plume cannot be made even with respect to arcjet nozzle position, so scanning through the plume in each axis and taking fluorescence data will be required to determine plume center and nozzle exit location. 17 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Velocity measurements of the plume are taken with respect to the laser beam's orientation. To this end two laser paths are installed so as to measure axial velocity (beam shining into arcjet nozzle) and radial/azimuthal velocities (beam shining perpendicular to nozzle exit normal vector). Care needs to be taken that these beams are in fact parallel and perpendicular to the thruster nozzle exit plane. 2.2.1.2 Laser Power Due to the fact that the techniques mentioned above are very sensitive to laser power, strict control over laser power will need to be maintained. Power measurements will need to be made regularly between scans to verify constant laser power and pulse-to-pulse measurements during the scan will be used to calibrate intensity data. Laser power meters will be required to monitor the pulsed laser operation and photodiodes in the ultraviolet range will be required to more accurately measure pulse to pulse fluctuations during laser wavelength scans. In order to conduct power dependence studies of the fluorescence signal with respect to the laser power, a high-power beam variable attenuator block will be required to reduce the laser signal without changing the beam shape to avoid changing the size of the collection volume. Power dependence studies that utilize this attenuator will investigate saturation broadening, ASE, and the power-squared fluorescence response that is expected for a two photon excitation scheme. These issues are discussed in more detail in following sections. 2.2.1.3 Detection o f Fluorescence Detection of the fluorescence in the bright arcjet plume is a difficult problem and will be one of the driving factors on laser power. Increased laser power will increase fluorescence for a given 18 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. density of atoms in the probe volume and allow the signal to more easily be observed over the background radiatioa Increased laser power, however, will also lead to multiphoton ionization, ASE, and saturation, which must be avoided in order to accurately interpret the signal. To this end, increased effort in efficiently collecting the signal is required. For the sake of sensitivity and time response, a photomultiplier tube (PMT) that is sensitive in the general regions of detection (red through blue) is used. A gated PMT socket will be used in order to assist the response of the PMT so that it is not continuously registering signal from the bright arcjet plume background and is only operational near the time of fluorescence. According to the manufacturer (Hamamatsu) this gating operation can add up to an order of magnitude sensitivity in bright background environments by allowing cathode amplification voltage to remain higher during the time of operation than it would if it was in continuous operation. To most efficiently collect light from the arcjet plume, a lens is placed as close as possible (without it or its mount melting, not that this ever happened...) to the arcjet nozzle exit. The lens collimates the collected light and a second lens focuses the collimated light on to the PMT. Bandpass interference filters are placed in front of the PMT outside of the chamber. Signal from the PMT is either digitized by a digital oscilloscope for purposes of determining signal lifetime or laser pulse width, or is collected by a gated integrator to amplify and average the signal from shot to shot. 2.2.1.4 Resonant Scatter If the backup resonant detection technique had been used (the two step, three photon excitation scheme, for example) where detection takes place at the same wavelength as one of the laser beams, care needs to be taken to prevent laser scatter from entering the collection optics. 19 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Baffles and tunnels can be placed around the optical train to prevent stray light from entering the detector. Additionally, resonant scatter may prevent the diagnostic from taking data near nozzle exit as light may acmally scatter off of the arcjet nozzle from the laser beam and into the detection. This is of greater concern when the beam approaches the arcjet parallel to the arcjet nozzle and is not such a concern when the beam is perpendicular to the nozzle. When the beam is perpendicular to the arcjet, an appropriate beam dump is placed in the path of the beam a distance away from the collection volume to prevent reflection of stray laser light. As mentioned earlier, non-resonant excitation can easily be protected from laser scatter through judicious use of spectral filters in front of the detection. 2.2.1.5 Measuring the Collection Volume Density measurements depend not only on accurate understanding of fluorescence with respect to number density, but accurate understanding of the collection volume is also important. The collection volume is defined as the intersection of the laser beam cross section with the detection focus cross section. Determining the size of the laser beam at the point of the detection is important for this calculation and is accomplished by imaging the beam onto a photodiode near the collection volume with the laser source on the other side of the collection volume. A razor blade is passed through the collection volume attached to a motion control stepper motor system. By slowly moving the razor blade through the beam, the size of the beam from fully blocked to not blocked may be determined. Assuming a Gaussian laser shape, a beam waist can be determined from half-width of the trace of intensity vs. position. 2 0 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Additionally, a light source is placed in the PMT housing with the tube removed so as to image light to the collection volume back along the optical train. This can also be measured in the same fashion and the intersection of the two measurements defines the collection volume. 2.2.1.6 Positioning in the Plume To make spatial profiles of atomic density in the arcjet plume, the diagnostic is applied to several locations across the nozzle exit of the 1 kW hydrogen arcjet. Rather than moving the entire optical detection train and the laser optics, movement of the thruster itself on a three dimensional motion control axis is chosen. This allows the optics to remain fixed once aligned. Definition of the nozzle exit is designated as the point where signal drops to half of what it was before approaching the nozzle exit. Adequate laser power and spot size are both of concern, and a focusing lens is placed in the chamber for the laser when the laser passes perpendicular to the nozzle exit normal vector. For velocity measurements where the laser needs to fire into the nozzle and into the oncoming gas flow, a two lens telescope system is used to focus the beam down to as small a spot as is possible from outside the vacuum chamber. Lenses for this beam path are not put into the chamber as the heat from the plume in this direction is substantial. 2.2.1.7 Modeling the Raw Data To accurately differentiate density, temperature, and velocity measurements, Gaussian profiles are least-squares fit to spectral profiles at each spatial location in the plume where density is measured (The applicability of Gaussian profiles rather than Lorentzian or Voigt profiles are discussed in sections 2.2.2.1 through 2.2.3.2). This allows accurate determination of line centers, line widths, and line areas for use in the measurements mentioned. Each profile is fit to a Gaussian 2 1 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. formulation and sized to the measured data using a Levenberg-Marquardt least squares fit."° The wavelength shift of the center of the line profile, with respect to the line profile of a stationary gas simultaneously probed, yields the axial velocity when the laser beam focuses down the axis of the thruster into the nozzle, and the radial or azimuthal velocity when the beam is perpendicular to the nozzle exit normal vector. The width of the Gaussian profile yields the temperature of the gas probed. These measmements are discussed in greater detail on the sections discussing velocity and temperature measurements (sections 2.2.4 and 2.2.3.) 2.2.2 DENSITY M e a s u r e m e n t I ssu e s Density measurements are made based upon accurate integration of the absorption lineshape of the transition that is probed. Fluorescence signals from different parts of the arcjet plume, and additionally a calibration environment, need to be compared for relative and absolute density measurements. This can only be successful if the integrated fluorescence in each area of the plume (and calibration cell) is understood with respect to the appropriate broadening mechanisms and artificial depopulation mechanisms taking place at different spatial locations in the plume. Mechanisms that carmot be accurately accounted for must also be avoided. 2.2.2.1 Broadening Mechanisms One broadening mechanism to be addressed is saturation broadening due to high laser energies. This broadening of the line occurs when the line center becomes saturated and the wings increase while the center of the transition absorption line does not with increased laser energy. This mechanism can be observed (along with other mechanisms described below) when the integrated fluorescence signal deviates from the power-squared relationship between laser power 2 2 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. and total fluorescence present. The fluorescence should follow a power squared relationship to laser power due to the two photon excitation. A power dependence study varying laser power and measuring total fluorescence is conducted to determine the laser power range acceptable to insure that a non-saturated transition is being probed. A small amount of line broadening is caused by the line width of the dye laser beam which, unlike that of continuous wave ring dye lasers, is non-negligible compared with the atomic linewidth. The laser linewidth of the available dye laser when tripled to 205 nm (for the two photon direct technique) has been measured by taking an LIF spectrum of NO gas in a calibration cell. Since NO is much heavier than hydrogen, and is at room temperature, the width of the rotational lines in the LEF spectrum reflect only the linewidth of the laser (as long as care is taken to use unblended rotational lines). This has shown that the available laser has a width of 0.28 cm ’ or 0.012 Â at 205 nm. The effect of the laser width in a two phton fluorescence measurement is taken into account using the following relation:"’ Equation 2-11 Av = ^AVg 4- 2Av7 where Av^ is the true Doppler width of the line (in cm ’), Av, is the width of the laser, Av is the measured transition width, and the factor of two is due to the two-photon probe method. Stark broadening mechanisms are present in the arcjet plume and need to be understood and quantified. Stark broadening is not expected to affect the density measurements significantly and is expected to be of greatest concern for temperature measurements where the temperature is determined as a function of linewidth squared. A more detailed discussion on the effects of Stark 23 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. broadening is given in section 2.23.2 as it relates to the overprediction error in measuring temperature due to unknown electron densities. When looking at the total convolved lineshape, the Doppler broadening component (as is demonstrated in the following sections) dominates, and the Stark component is minimal. If the laser linewidth is deconvolved from the total linewidth prior to modeling, and the laser power is kept low enough to avoid saturation, a Gaussian profile with the appropriate uncertainty due to Stark effects appears to accurately model the transition linewidth. 2.2.2.2 Collisional Quenching The arcjet plume environment has substantial density gradients as the plume expands from the thruster nozzle exit. The changes in density with respect to position result in changing effects of depopulation of excited state hydrogen atoms due to collisional quenching. Depopulation due to quenching prevents collection o f fluorescence from all promoted electrons that are related to the density of the atomic ground state hydrogen that is present, and a correction for the quenching rate at the location of the measurement must be made. One method of determining the quenching correction is to determine the effect of quenching through observation of the lifetime of the fluorescence signal on a fast 500 MHz digitizing oscilloscope. The lifetime of the transition can be determined by finding the 1/e time of the exponential decay of the fluorescence observed The measured signal decay is a convolution of the laser pulse decay and the transition decay and is deconvolved prior to the measurement. Relating the lifetime at each point in the plume to the unquenched lifetime gives a linear relationship correction factor for the measured integrated fluorescence at that point. The effective unquenched 24 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. signal is then determined and corresponds to a relative density value. Calibration of the signal is required to convert the relative density measurement to an absolute number density value. 2.2.23 Saturation, Multiphoton Ionization, and Amplified Spontaneous Emission A focused pulsed laser beam creates a very high instantaneous energy density, which is needed to excite the two-photon transition, but also can induce other non-linear processes. Multiphoton ionization (MPI) can occur, and partial saturation of the transition becomes possible, even though the two-photon absorption cross section is very small. In addition, amplified spontaneous emission (ASE) can take place for the direct two photon excitation technique, since a population inversion is created between the pumped n=3 level and the nearly empty n=2 level. The population inversion can cause gain to occur for any photon emitted in the forward or backward direction along the laser beam.'^"^^ If ASE or MPI become large, they become a non-negligible loss mechanism for the LIF process. Saturation causes power broadening of the Doppler profile of the transition, and saturation, ASE, or MPI can cause the power dependence of the LDF signal to deviate from the expected behavior proportional to power squared. 2.2.3 Te m p e r a j u r e M e a s u r e m e n t I ssu e s Temperature measurements are based upon accurate determination of the Doppler line broadening due to thermal gas motion. Understanding the component of linewidth directly associated with Doppler broadening is essential for temperature determination. 2.2.3.1 Doppler Broadening Temperature is based upon the square of the measured Doppler width and can be written in terms of width in the following maimer:'® 25 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Equation 2-12 Ô X . = 7.16 x lO'^X. where SX is the linewidth due to Doppler broadening, X o is the line center, M is the molar mass of the gas, and T is the translational temperature. The constant in the equation is based upon the values of the speed of light and the molecular gas constant. Note that when solving for the temperature as a function of the Doppler linewidth, the width is squared and any error associated with the measurement of Doppler width is also squared. 2.2.3.2 Stark Broadening One broadening mechanism that is present, and must be understood to determine accurate Doppler line widths, is Stark broadening of the desired atomic hydrogen transition due to the free electrons present in the arcjet plume. A direct measurement of the profile of the electron number density for this arcjet was not undertaken, but electron number densities have been measured for !3 -3 ^ nearly identical conditions to be less than 2 x 10 cm at the nozzle exit plane. This maximum value of ne would cause a Stark width of 0.002Â for the L(3 line.^ A simulated Voigt profile using this Lorentzian width shows that, for a typical measured linewidth of 0.029Â, accounting for the Stark broadening would cause the temperature from the Doppler portion of the linewidth to go down from 1600K to 1490K. This represents a likely maximum uncertainty in the temperature due to Stark broadening, and is greatest at the center of the nozzle where ng is reported to be largest. 26 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2.2 .4 VELOCITY MEASUREMENT ISSUES Velocities are measured through determination of the wavelength where atoms speeding toward incoming photons absorb those photons at shifted wavelengths. This takes place due to the high velocity of the gas and the resulting apparent frequency shift. Tuning a laser near the wavelength where absorption takes place results in a spectral fluorescence profile centered at the shifted wavelength. As the wavelength shift is determined by relationship to the incoming photons, velocities are measured with respect to the direction of the incoming beam. Therefore, if the beam propagates axially into the arcjet nozzle an axial velocity measurement can be taken. If the beam approaches the nozzle from a radial direction both radial and azimuthal velocities can be measured depending upon position of the intersection of the beam and the collection optics with respect to the centerline of the nozzle. 2.2.4.1 Doppler Shift Velocity is measured by observing the Doppler shift of the absorption line and determining the velocity relative to the incoming photons responsible for the shift. Equation 2-13 v = ^ ^ c where v is the velocity, AX is the wavelength shift, X o is the zero velocity line center, and c is the speed of light. Determining the line center of a profile with the Gaussian fitting process is relatively simple. Performing the procedure on measurements simultaneously collected from both a static gas in a calibration cell and in the arcjet plume, provides a shifted Gaussian in the arcjet 27 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. plume and a calibration of zero velocity in the calibration cell. Determination of the shift between the two profiles, AÀ, allows the use of the relationship above to determine gas velocity. In arcjet plumes, velocity of species has been seen to vary from 5 km/s to 15 km/s using the above technique with accuracy as high as 50 m/s with continuous wavelength lasers and accuracy somewhat less resolved with pulsed lasers. In this case resolution will be a factor of six less than typical LIF velocity measurements which probe the excited state at 656 nm, as the transition being probed has an effective transition wavelength of 102.5 nm. Even with this constraint, velocity measurements of the ground state should still be quite resolved with resolution expected to be about 500 m/s. 2 .3 C a l ib r a t io n f o r A b so l u t e D e n sit y M e a su r e m e n t s 2.3.1 A TOMJC H yd r o g e n C o n c e n t r a tio n s in a Ca u b r a t io n C e ll The procedure for obtaining absolute number density is based on a method reported by Meier et al.'^"° After a relative number density scan and corresponding lifetime data are taken, the arcjet is then turned off and translated away from the collection volume, the chamber is opened, and a second calibration cell is placed in the detection volume with the laser beam passing through it. The LEF signal and lifetime is measured for the hydrogen atoms present in the cell. Atoms are present due to the flow of hydrogen gas through a microwave discharge prior to entering the cell. Since the laser beam, detection optics and electronics have remained the same, the LDF from the arcjet and from the cell wül have the same proportionality to absolute number density after correcting for any differences in quenching: 28 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. gA Equation 2-14 -C, and s'" are the signal sizes (integrated over the entire spectral width) from the arcjet and the cell, and are the absolute number densities of atomic hydrogen in the arcjet and the ceU, and Cq is the scaling factor for the difference in quenching. is given by the ratio of fluorescence decay rate in the arcjet to that in the calibration cell. The absolute number density of hydrogen atoms in the discharge cell is obtained using a standard chemical titration method. A schematic of the calibration cell used for this method is found in Figure 2-5. The cell is made out of glass tubing and is approximately 30 cm (12 in.) long and 5 cm (2 in.) in diameter. Important to the design of the cell is the long drift tube area with a Teflon™ tubing liner to reduce gas/wall interactions. This drift tube allows complete mixing of the titration gas and the gas of interest. In this case, the titration gas is NO; and H is the gas to be calibrated. 29 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Optical Detection Focua ^ 12* long Glass Tube with Teflm Liner m sdng legiao Pressure Tap HyTle mixture eate n (hifl tube after passing through microwave discharge To Pump Window Lens 203mn Laser Beam Figure 2-5: Density Calibration cell schematic. The variability of large numbers of wall collisions in the drift tube can potentially cause great difficulty in the calibration, and the walls of the calibration cell are therefore coated with Teflon™ in order to reduce H depopulation due to the walls. Measurements of H atoms at different locations in the drift tube, with and without the liner, show a remarkable difference. With the Teflon™ liner in place, collisions with the walls are significantly lower to the point tfiat changes due to the location in the drift tube where the measurements are taken (i.e. the amount of wall area available for collision to the H atom) became almost imperceptible. Once the collection volume is fixed in the cell, any reduction in H atoms due to the addition of NO; appear to be reductions due to the chemical processes without large contributions due to wall collisions. The hydrogen (with a helium carrier gas) enters the drift tube area near one end after passing through a microwave discharge. A dilute mixture of NO: in helium is added to the flow through a 30 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. long thin tube and enters the larger drift tube region at a location past the point where the hydrogen has entered. This reduces the amount of NO; that might diffuse upstream and enter the microwave discharge. Placing the end of the tube somewhat closer to the collection volume than the microwave discharge leads to a better response of signal to NO; addition. This is believed to be due to reduced NO; upstream diffusion which, if allowed to become significant, appears to influence the number of H atoms produced in the microwave discharge. Placing the end of the tube five or six inches downstream of the entrance of the atomic hydrogen appears to prevent this effect for the following flow rates. The NO; reacts rapidly with hydrogen atoms when the two are completely mixed in the large drift tube volume. The calibration involves flowing a 2% hydrogen in helium mixture and then the addition of 2% NO; in helium until the fluorescence signal decreases. A small vacuum pump brings the cell pressure to about a hundredth of an atmosphere during this procedure. The calibration involves observing the LEF signal decrease until it is gone at the point where the partial pressure of added NO; is equal to the partial pressure of hydrogen atoms in the cell. Control of the amount of NO; added allows the determination of the amount of hydrogen atoms present in the calibration cell. The NO; concentration at the point the fluorescence disappears is equal to the concentration of atomic hydrogen present as each hydrogen atom is reacted away for each NO; molecule added in the fast reaction: Equation 2-15 H+NO; — > OH+NO The reaction above has a low activation energy and is very fast at room temperature. The large amount of buffer gas (in this case He due to availability, but could be Ar, or another 31 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. available inert gas) and low percentage of H significantly reduce possible secondary reactions, making them negligible for the purposes of this experiment. Calibration takes place either before or after each arcjet firing. 2.3.1.1 Chemical Reactions in the Cell The titration approach based on Equation 2-15 makes the assumption that the reaction with NO: is the sole reason for the depopulation of H atoms when NO: is added Possible recombination of the two radicals, atomic H and the OH product, for example, requires a three body collision to take place as shown in Equation 2-16 and is, therefore, significandy less frequent then the fast reaction in Equation 2-15. Equation 2-16 H-t-OH-t-X H:0+X Equation 2-17 OH-^H:— > H:0+H Equation 2-18 OH+OH— > H :0+0 The repopulation of H, after Equation 2-15 takes place, through processes such as the one shown in Equation 2-17 are found to be negligible due to energetics and the high activation required for Equation 2-17 which is essentially not available at room temperature. Additionally, the probability that the two minor species radicals will combine as in Equation 2-18 is very low and this process can not compete with the faster H and NO: reaction. Because of these factors, the assumption that Equation 2-15 is by far the dominant reaction and the cause for reduction of H atoms as NO: is added in tfüs titration approach appears to be a good assumption and is therefore supported by many researchers.'®"^ 32 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2.3.2 COMPARING Ca UBRA TION CELL AND ARC JET DENSITY MEASUREMENTS Once a relative density profile has been taken as described in section 2.2.2, and quenching, ASE, MPI, laser linewidth broadening, etc. have been compensated for, or taken into account, an effective fluorescence measurement results, representing relative density in the arcjet plume. Through the method described in the previous section, a relationship between H atom density and a corrected effective LIF fluorescence for the calibration cell environment may be determined. If the correction factors in each environment are appropriately made so as to bring the fluorescence measured in each environment to a standard “ideal” condition equivalent, a relationship between the fluorescence measured in the arcjet plume and the corresponding atomic hydrogen number density may be made. 33 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 3 Experimental Setup PPU Ttmeble Dye Aiqi PMT & PeQizKBroa Ptôm PMT Leo» PMT FOtov PMT Figure 3-1: Experimental setup. 3.1 O v e r v ie w A diagram showing the experimental setup is shown in Figure 3-1. A NASA standard, laboratory model IkW arcjet thruster^' operating on hydrogen propellant is run in a vacuum chamber located at the Electric Propulsion Laboratory of the Phillips Laboratory at Edwards Air Force Base. The arcjet operates on a hydrogen gas flow of 13.1 mg/s (8.74 slpm) at an operating chamber pressure of 6 Pa (45 mtorr). The gas flow and power are chosen to closely match the conditions under which the most complete set of previous diagnostic data for a 1 kW hydrogen 3 4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. arcjet is taken, in order to facilitate comparisons between different data sets and between data and models. The laser is a pulsed dye laser pumped by a Nd:YAG with a repetition rate of 10 Hz and a pulse width of 6 ns. For the two photon direct excitation method, the dye laser output is at 615 nm and is frequency tripled using beta-barium borate (BBO) crystals to achieve about 0.5 mj per pulse at 205 nm. A mirror turns about 80% of the beam toward the arcjet chamber, but first passes through a variable attenuator placed in the beam. The remaining 20% of beam energy is directed toward a microwave-discharge source of atomic hydrogen. For axial velocity measurements, the beam is sent directly down the axis of the arcjet flow (Path I ) and is focused with a telescope lens configuration (not shown) outside the chamber with a focal length on the order of 2 m. For radial measurements, the unfocussed beam is sent to a turning prism inside the chamber located underneath the arcjet (Path 2). directed to pass vertically through the plume, and is focused by a 200 mm lens. The laser beam and optics remain fixed, while the arcjet is mounted on a motion control x. y. z stage to translate it for probing different regions of the plume. A filtered photomultiplier tube (PMT) is placed behind the last turning mirror before the beam enters the chamber. This is to detect amplified spontaneous emission (ASE) that may propagate back along the laser beam path originating from inside the arcjet plume. A 200 mm focal length, 2" diameter lens is placed inside the chamber to collimate the LIF signal that is emitted toward the side window. The light is collected outside the cfiamber, focused through a 1 mm aperture, and detected with a filtered, gated PMT. Since the LIF occurs at 656 nm, the filters used are a 656 nm bandpass interference filter and an RG 645 color glass filter (thus all scattered laser light (205nm) is filtered out). The gated PMT is an ordinary Hamamatsu 928 tube with a specially provided 35 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. socket from Hamamatsu that is triggered to detect light for 2 ms during each laser pulse, and is off between pulses. This is an important feature that allows operation of the PMT at full voltage without exceeding the anode current limit due to the bright arcjet emission background. Saturation is avoided by setting the voltage level so that both average current and peak current during the gate remain below the maximum specified for the PMT. A gated integrator with a 30 ns gate is used to amplify and average the Hoc fluorescence seen by the PMT. Alternatively, the PMT signal is also digitized using a fast oscilloscope to obtain fluorescence lifetimes and quenching information. The weak UV beam that is sent into the chscharge cell is focused with a 150 mm lens. The cell is run with a slow flow of a few torr of helium carrier gas and a few percent hydrogen through a microwave discharge tube. The LIF in the discharge cell is detected through a filtered (ungated) PMT. Simultaneous detection of LEF from the cell during each spectral scan of the arcjet LIF provides a zero-velocity comparison from which to measure Doppler shifts and room temperature Doppler widths. This discharge cell provides calibration in wavelength space but is not adequate for density calibration since the density of Hydrogen atoms is not known in this discharge cell, and because the detection equipment and optical path differs from the setup on the arcjet chamber. A separate discharge cell is used for density calibration and is placed inside the chamber when the arcjet is not operating. Using the above experimental setup, density, temperature and velocities (axial, radial, and azimuthal) of ground state hydrogen atoms are measured in the arcjet plume. 3 .2 L a s e r E q u ip m e n t The following sections discuss the laser equipment used in this experiment. Lasers, trackers, and other apparatus are presently required to create beams of photons at the powers and 36 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. frequencies required for this experiment. It is expected that as laser technology progresses, a more simple setup may become possible. 3.2.1 N d. YAG PULSED L a s e r A Continuum brand NY81C-10 pulsed Nd:YAG is used in this experiment to pump the tunable dye laser in this setup. This laser produces energy for about 6 ns at a 10 Hz repetition rate and is set to produce light at the 532 nm wavelength using internal Type II doubling. At this wavelength, the laser is specified to be able to provide 550 mJ of energy per pulse. Average observed energy is closer to 350 to 450 ml per pulse depending upon the day. The laser requires external cooling of 3 to 4 gallons per minute and has an internal heat exchanger to maintain the purity of the internal laser coolant. The linewidth of the laser is not an issue in this experiment since it is pumping the dye beam, but with the included injection seeded SLM module (which only worked for a short time before requiring maintenance) the linewidth is specified to be as low as 0.0045 cm '. The rod diameter of the laser is 9.5 mm and produces a symmetrically circular beam (as observed by imaging the beam onto flash paper). 3.2.2 T u n a b le DYE L a s e r A Continuum brand ND60 dye laser is used as the tunable laser source and is pumped by the NY80 series Nd;YAG laser described above. The ND60 contains a single oscillator and two amplifiers. The last amplifier before exiting the device is of a capillary design and is internally 5 mm in diameter. The dye cells in the laser are transversely pumped and energy is divided such that 5% of the pump energy goes to the oscillator. 10% goes to the preamplifier, and the remaining 85% pumps the final amplifier. The oscillator wavelength selection is achieved by the adjustment of a 37 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. grating and mirror cavity combination where the mirror rotation is controlled by a high resolution stepper motor and the grating remains fixed. To achieve the output beam at 615 nm, a suiforhodamine 640 dye (purchased from Exaton) is used with 214 mg/1 concentration in the oscillator and 110 mg/1 in the amplifiers. The peak wavelength of the dye is at 613 nm and the dye is specified over a range of 605 nm to 630 nm. 3.2.3 Wa v e l e n g t h D o u b u n g & T r ip u n g The wavelength of tunable nanosecond pulsed lasers can be extended into the ultraviolet by the use of second harmonic generation, followed by additional third harmonic generation. Generation of these harmonics can be accomplished by sending the beam through the appropriate crystal types such as BBO crystals and Potassium dihydrogen phosphate (KDP) crystals, depending upon the laser source and wavelength range. For a fixed wavelength, optical setup of the harmonic generation is not difficult, but when generating harmonics on a tunable beam source, an electronic servo driver is required in order to maintain the phase match angle with respect to the incident wavelength as that incident wavelength changes. A feedback and control loop analyzes the signal from the crystal and maintains it as the source wavelength is tuned in this scenario. This servo driver is required for both the doubling and the tripling crystals in this setup. Two INRAD Autotracker II electronic servo crystal mounting devices and controllers are used in order to triple the near 615nm light coming from the tunable dye laser. BBO “A” crystals from INRAD are used with CB-41 compensator optics to first double the 615 nm light to 307.5 nm and then triple the light to 205 nm No color filters are required for this setup, though neutral density filters to protect the feedback-control mechanism are used liberally. 38 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. One of the requirements of generating short wavelengths in third harmonic generation is that the two input wavelengths must have polarization vectors aligned in identical directions. To satisfy this requirement, a polarization rotator designed for nanosecond lasers is placed between the doubling and tripling crystal mountings. A model BC-BHl 100 rotator from INRAD is used in the experimental setup between the two Autotrackers. 3 .2 .4 Wa v e l e n g t h S e p a r a t i o n The output of the third harmonic generating Autotracker contains the precious 205 nm photons that are required for this experiment. Unfortunately, the beam also contains the 307.5 nm light, the 615 nm light, and perhaps some scattered 532 nm light from the pump. In order to separate the 205 nm light from the other photons, a pellin-broca prism is used to disperse the photons as a function of wavelength. All beams but the 205 nm beam are then directed into beam stops. A 4" by 3” prism is used to maximize the spatial dispersion of the different wavelength beams, but minimize the amount of travel the 205 beam spatially takes as the local tuning takes place over the transitions of interest. 3 .2 .5 E n e r g y A t t e n u a t io n In this experiment, generation of the 205 nm photons requires significant laser power in order to overcome the losses at each step in the process. Additionally, control of the laser power is required in order to prevent physical processes such as multiphoton ionization and amplified spontaneous emission from overtaking the fluorescence and preventing the collection of data related to density and temperature. Ultraviolet beam attenuation is accomplished with a Newport 935-5 series high power attenuator. Using Fresnel reflection from four uncoated, counter rotating fused 39 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. silica wedges, the attenuator has an inherently high power handling capability over the 200 nm to 2.1 micron wavelength range. One of the advantages of this device is that because of the multiple optical elements used in attenuation, output beam deviation is negligible over the range of attenuation used in this experiment, allowing a fixed optical path downstream of the attenuator. A micrometer control is used to set the anenuation, and energy meters are used to verify percent reduction in signal. 3 .2 .6 m e a s u r e d L a s e r L i n e w i d t h o f F in a l b e a m As mentioned in section 2.2.2.1, the laser linewidth of the available dye laser when tripled to 205 nm (for the two photon direct technique) has been measured by taking an LIF spectrum of NO gas in a calibration cell. Since NO is much heavier than hydrogen, and is at room temperattu’ e, the width of the rotational lines in the LEF spectrum reflect only tlie linewidth of the laser (as long as care is taken to use unblended rotational lines). This has shown that the final laser beam has a width of 0.28 cm ' or 0.012 Â at 205 nm. 3.3 B eam P o w e r /E n e r g y M e a su r e m e n t Several energy and power meters are used, calibrated, and compared to insiu’ e that the laser energy levels are consistent throughout the experiments. Measurements are made at key locations on the optical bench, between the Nd:YAG laser and the dye laser, at the output of the dye laser, at the output of the pellin-broca prism, after the turning mirror for the calibration cell, and both prior to and after the UV attenuation of the beam going to tfie test chamber. A Molectron Powermax laser power meter is used to determine average energy output from the Nd:YAG and allow adjustment of the pump laser to optimize its output to the dye. Two 40 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. separate joule energy meters from Laser Precision (model RJP-734) are used to measure average pulse energies with a IC X ) or 2(X ) pulse sample being typical for determining average laser energy. Additionally, temporal measurements of laser pulses are measured using Schottky-type GaAsP photodiodes (Hamamatsu model G 1126-02) sensitive in the ultraviolet light range down to 190 nm. Measurements of beam size at the thruster location also used these photodiodes, which were protected by high attenuation and blocked by a movable razor-blade attached to a motion control system. 3.4 M ic r o w a v e D isc h a r g e C a v it y fo r C a l ib r a t io n C ells Calibration cells used for wavelength and density calibration have been described in detail in previous sections. In both cases, the creation of H atoms in the discharge cell takes place when traces of H; gas mixed with He are sent through a microwave discharge cavity to create some number of hydrogen atoms which can be investigated in a more controlled environment than in the plume of the arcjet. In the case of the wavelength calibration cell, the atomic hydrogen is used to observe linewidths and wavelengths of essentially static, room temperature gas. This is compared to the linewidths and wavelengths of the atomic hydrogen measured in the arcjet plume. In the case of the density calibration cell, titration using small amounts of NO; is used in order to determine the total fluorescence, and hence, the density, of the H atoms produced by the microwave discharge cavity. In both cases the same microwave discharge cavity and supply is used. An Opthos M PG ^ microwave power generator produces up to 100 watts of power at 2.45 GHz. This microwave energy is connected to a Evenson style microwave reflection cavity positioned around 16" glassware using a flexible microwave wave guide and is ignited by the application of a discharge to 41 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the glassware from a hand-held portable Tesla coil. Once ignited, significant adjustment of the reflector cavity can be required in order to maximize the stability of the discharge and minimize the harmful reflected energy back to the generator. After about 30 minutes another adjustment is made and the discharge (in this semp, at least) remains steady fw the remainder of the day. 3 .5 G a t e d P h o t o m u l t ip l ie r D e t e c t io n As described in the overview section, detection of fluorescence in the control volume of the arcjet plume and calibration cell is accomplished using a Hamamatsu 928 tube with a special gated Hamamatsu socket (model C 1392-09) that allows the fluorescence to be detected in the presence of the bright background plume radiation by only allowing detection during the time the programmed gate is applied (which corresponds to the time the laser is triggered and LEF is taking place). When the detection is not desired, the socket blocks voltage to the initial dynode pair and prevents amplification of electrons emitted from the photocathode. If the dynodes are operational in the presence of bright emission, the amplification of electrons increases to the point that the dynode chain heats up and becomes nonlinear in response to input signal. Additionally, damage to the photocathode takes place in the presence of a significant number of incoming photons, not so much because of the photons themselves, but rather because large amounts of amplified electrons at the end of the dynode chain end up ionizing the trace amount of gas remaining in the PMT. These charged ions are then drawn to the photocathode and the impact of these particles with mass and potential cause permanent damage to the photocathode from the back interior surface. Since the gated socket prevents the electron amplification by blocking the dynode chain it both prevents the chain from becoming nonlinear during the time between laser pulses and protects the photocathode from internal ion damage. 42 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Determination of the appropriate amplification level, so as to avoid saturation and damage to the PMT during signal collection, is accomplished by periodically using a Keithley Metrobyte picoameter on the output of the PMT. Using the specification from Hamamatsu of 100 microamps at 1250 volts amplification as maximum average anode current at maximum amplification voltage, tfie output signal is monitored over a period of time where many open/close cycles take place. By keeping the amplification voltage low enough to insure tfiat the output signal is beneath maximum output ciurent, but high enough to distinguish LIF signal, a setting for PMT amplification is determined to be around 1100 volts for this experimental setup. Through use of neutral density filters placed in front of the PMT at that amplification, linearity of response to intensity was demonstrated, insuring tfiat the PMT would behave in a linear fashion at tfiis amplification voltage. 3 .6 A r c j e t S y s te m The one kilowatt class arcjet thruster tfiat is used for tfiis experiment was designed by NASA Lewis Research Center^' and was part of a program where identical models of arcjet thruster and power electronics were supplied to a variety of research institutions so tfiat many different locations could participate in research on this technology. Tfiis tfiruster and power supply were provided for tfiis research on a renewable contract to the University of Southern California. As shown in Figure 3-2, tfiis arcjet model has a center cathode electrode and an outer anode electrode which is shaped in the configuration of a converging/diverging nozzle. Propellant (in tfiis case, hydrogen gas) enters the tfiruster and is directed to swirl around the cathode and then pass through the arc which attaches to the anode downstream of, and tfirougfi, the nozzle tfiroat. In tfiis region, the gas is heated to temperatures as high as 20,000 °C near the cathode tip and then expansion occurs tfirough the nozzle to form directed tfirust. 43 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. I Anode Propellant Figure 3-2: Simple Arcjet Cutaway Diagram The propellant flow rate is controlled and measured using an MKS mass flow controller (series 1100) calibrated for use on hydrogen. Additional calibration was conducted in the laboratory using two independent techniques. A wet test gas bubble meter was used over the course of several hours to determine the flow rate vs. flow controller setting curve. Fortunately, in the range of propellant control used for this experiment the controller provided a linear response that was easily converted to mg/s. An additional verification took place using a fixed volume. Pressure, temperature, and flow rate were measimed over a fixed time to determine the amount of hydrogen that went through the flow controller and into the fixed volume. Results from this test correlated well with the wet test gas bubble meter results. Throughout the experiment the pressure in the propellant feed line is measimed with an overpressure MKS Baratron 122B inside the vacuum chamber near the propellant entrance to the arcjet. A reading of just over I atmosphere was observed throughout the testing. Another two MKS Baratron 122B’s calibrated for measurements down to 10'^ torr are used for vacuum chamber pressure measurements. Both 44 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. transducers were in agreement and were additionally correlated with MKS pressure thermocouple gauges. The arcjet is electrically isolated from its mount and from the entire chamber. Because the propellant line of the arcjet is attached to the anode, the propellant line requires isolation as well. A custom propellant line spacer made out of Alumina with swage ends welded on each end of the spacer was manufactured at the Jet Propulsion Laboratory’s Advanced Propulsion Lab’s machine shop. This spacer allows the propellant line in the chamber to be isolated from the arcjet propellant swage fixture which floats at the arcjet anode potential. 3.7 T h r e e -A x is S p a t i a l A r c j e t P o s it io n in g S y s te m In order to probe the arcjet plume region without adjusting the laser and collection optics simultaneously, the arcjet itself is moved in the vacuum chamber on a three axis positioning system using a two dimensional positioning table and a vertical stage orthogonally attached to the table. All motion stages and fine stepper motors are produced by Daedal/Parker systems. The motors used had 1000 steps per revolution and the thread pitch on the stages are 10 revolutions per mm allowing a single step command to move the arcjet as little as 0.0001 mm in any direction. In practical use, the smallest amount the arcjet is moved in order to notice changes in signal is 0.005 mm, with average movements of 0.05 mm quite common. Alignment of the thruster on the positioning table is completed through the use of optical alignment and signal detection off of the tip of a metal rod attached near the arcjet to the motion control system. An alignment laser or the 205nm beam can be focused onto the fine point of the metal rod and either the reflection of the alignment laser or the sparking off of the tip of the metal due to the 205 nm beam can be detected using the collection optics. Many hours climbing inside of 45 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the chamber to insure that the proper alignment was taking place through the chamber door window (with the door in its closed position) has been gleefully accomplished. Additionally, a computer record of motor position is continually kept and used for alignment verification. Once the thruster is lit. very small adjustments in alignment take place based upon initial plume velocity measurements conducted prior to large amounts of data collection. The zero position of the radial velocity is easy to determine and found to be highly repeatable and reliable. The axial zero position is determined using the front of the arcjet nozzle by determining the point at which the fluorescence signal drops to half in the collection volume due to the fact that the nozzle body fills half of the small collection volume. 3.8 D a t a A c q u is it io n E q u ipm en t Getting the laser to produce 205nm light, tuning the laser on demand, getting the crystals to track the wavelength, aligning the laser and the collection optics, adjusting the signal level out of the PMT, calibrating the laser wavelength, attenuating the laser to provide different fluorescence signals to determine power squared response, calibrating the signal intensity in the presence of a determined amount of H atoms, installing and removing the density calibration ceU in the confines of the chamber, moving the arcjet to the correct position, getting the arcjet to work, the chamber at low vacuum, the propellant flow calibrated, and finding the appropriate Doppler shifted transition at a location in the plume is nontrivial, however, it is still not enough. The data has to be collected, stored, and analyzed, often immediately, so that a decision on the next data point to take can be made, and a determination on the confidence in the data being taken can take place. The following section provides an overview of the data collection apparatus. 4 6 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Current from the output of the photomultiplier tube is terminated in 50 ohms for fast response and input to a Stanford Research Systems SR250 Gated Integrator and Boxcar Averager. The unit is triggered in concert with the laser beam and delayed until the fluorescence is measured. In this way, a “window" of signal detection is measured and averaged with the previous measurement on a sliding exponential scale. Because of the laser frequency, ten measurements are taken per second. Amplification of the signal measured during the gate time also takes place, and the signal level is continuously output and sent into an analog to digital channel of a National Instruments high resolution AT-MIO-16X I/O board for the IBM-PC compatible computer. As the laser is tuned over the transition wavelength in order to determine lineshape, the signal into the computer provides the fluorescence resulting from the wavelength chosen. In order to support a large sample size for each measurement, laser scan rates take place on the order of 0.0005 angstroms per second. A scan over 0.15 angstroms in wavelength space at a single spatial location took approximately 5 minutes. One second averages (averaging = 10) are typically used on the boxcar ager. Because multiple boxcar and I/O channels are available, simultaneous measurement of the plume PMT during the fluorescence, the wavelength calibration cell PMT, a photodiode measuring laser intensity, and the plume PMT measuring only the background signal are collected so as to provide a background subtracted signal which can be much more easily handled by the computer. At some points in the experiment, an ASE signal behind the last turning mirror was also collected to insure that the laser power was below the threshold where significant ASE could be detected. At times when the fluorescence lifetimes are to be measured, the signal is not routed to the gated integrator and boxcar averager, but rather sent to a high speed storage digital oscilloscope. The Tektronix 644B oscilloscope provides 8-bit 1 GHz bandwidth digitizing with 2.5 47 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Gigasamples/second sampling rate on each of its four input channels. This allows detailed sampling of the -10 ns lifetime of the hydrogen relaxation and the input laser pulse (as measured by the photodiode). Transfer of the data stored on the oscilloscope to the computer is accomplished using a GPIB or IEEE 488.2 connection between the scope and a National Instruments AT-GPIB communications board. The timing of all of these collection and laser events is quite important to the experiment. A Stanford Research Systems Model DG535 Digital Delay Generator is employed to trigger the laser, the four boxcar averagers, the PMT socket, the computer data sampling, and the laser detection equipment (energy meters and photodiode) to ensure accurate detection and collection timing. The control of the analog data collection, the GPIB waveform downloading, the arcjet position location (via the stepper motor motion control system) and the laser wavelength and scan rate was all accomplished using an IBM personal computer and Lab View software. Communications with the laser took place over the RS-232 serial port, while the AtoD and GPIB communications took place with the additional expansion cards mentioned above. A control card for the motion control system was also placed into an available slot on the computer. The software used could set the start laser wavelength, end laser wavelength, arcjet position, and scan rate, and then collect the data from many different channels. Waveforms from the oscilloscope were also collected and stored. Data analysis software (written in FORTRAN with Numerical Recipes''^ libraries) was used to fit lineshapes and determine exponential decay of lifetimes then used the collected data as input, and would provide momentary results based on the collected data. Detailed alignment and data collection without this instantaneous response would have made this experiment far more unwieldy than it already was, and perhaps beyond the scope of completion. 48 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 Results and Analysis Fluorescence measurements of the atomic hydrogen in the arcjet plume are made using the techniques and apparatus previously described Taking the light collection and drawing conclusions on atomic velocities, temperatures, and densities in the arcjet plume, requires several calibrations and secondary measurements. The raw data and incremental analysis are given in the following section to elaborate on the process of transforming the raw measured data to the physical parameters that are reported. 4.1 DATA VERIFICATION 4 .1 .1 D e t e c t io n o f F l u o r e s c e n c e a t Va r y in g L a s e r W a v e l e n g t h s Figure 4-1 shows a representative wavelength scan involving simultaneous two photon excitation of atomic hydrogen in a calibration cell and excitation of atomic hydrogen at a location 0.4 mm away from nozzle exit in the arcjet plume, irradiated in the axial direction. A Gaussian least-squares fit (as described in section 2.2.1.7) overlays the data and is used for determination of physical properties of the atomic hydrogen species in the plume. The velocity is determined by measuring the wavelength shift of the center of the absorption line of the high speed gas plume with respect to the absorption of the light in gas without the additional energy of high velocity towards the incoming photons. The temperature is determined by the width of the absorption. Higher temperatures lead to a larger random velocity component that spreads out the wavelength of absorption for a statistical group of atoms. Additionally, the density is ascertained through the 49 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. determination of the number of absorbers which can be indirectly measured by comparing the total integrated fluorescence observed in the plume to that measured in a calibration cell. There are two calibration cells for this experiment and the data shown in Figure 4-1 is taken from the cell used to calibrate for temperature and velocity. 205.140 B 3 0.8 ■Ê 3 ^ 0.6 s dX S g c i e 3 0.4 fCalibratioi Cell Arcjet Plume 0.2 205.145 205.150 205.155 205.160 205.165 Laser Wavelength (nm) Figure 4-1: Sample 2PLIF spectrum of the hydrogen atom, showing the calibration signal from the discharge cell and the signal from the arcjet plume along with Gaussian least square fit profiles. Note that the fit of the theory to the data for the cell is extremely good as the environment of the static cell is at room temperatime and essentially zero velocity (when measured in km/sec). The atomic hydrogen in the cell is created by flowing gas slowly through a microwave discharge into a larger volume (at pressures of -1/300 of an atmosphere) where the measurement is made and 50 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. allows the atoms to be probed at velocities slow enough that no Doppler shift would be measured with the equipment used in this experiment. Temperatures in the static cell should be close to room temperature, but may be elevated due to the energy of the microwave discharge. A thermocouple measurement was made in the cell which indicated room temperature gas, but the temperature of the atomic species created in the discharge may have still been higher as the number of collisions prior to fluorescence may have been few. Since the arcjet temperatures were significantly greater than those measured in the cell, the nature of using the cell temperatures as a calibration between the different temperature measurements in the arcjet remained valid. The fit of the arcjet plume data is not as good as the fit of the data in the calibration cell, though the features of interest, the width and center of the absorption line, as well as the overall fluorescence area, do appear to match the data quite well. The scatter of the plume data, with respect to the theoretical Gaussian, is not surprising given the high gas velocity and the volatility of the moving arc inside the arcjet thruster. The Doppler shift of the arcjet plume absorption line relative to the static absorption in the cell is evident, as is the Doppler width increase due to the much higher temperature in the arcjet plume relative to the room temperature in the calibration cell. The wavelength shift of the two line centers yields the axial velocity as described earlier, while measuring the width of each Gaussian allows calculation of a temperature. The data (not yet corrected for any additional broadening mechanisms) corresponds to a temperature near 20(X)K and a velocity near 10 km/s in the arcjet plume. 51 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.1.2 DETECTION OF FLUORESCENCE AT VARYING LASER POWER The intent of the multiphoton laser technique is to measure the properties of the atomic hydrogen atoms without intrusively changing their state. The success of this intent requires confirmation that the technique does not alter the state of the atoms, and a study of the atomic fluorescence with respect to the input laser power was completed The energy of the beam can cause processes to occur other than the simple promotion of a ground state electron to the desired level. This is more extensively discussed in section 2.2.2. If the fluorescence of the hydrogen using this two photon technique is other than proportional to the square of the laser power, energy from the relaxation, or the laser itself, may be going into these other processes. Figure 4-2 shows the integrated fluorescence measured with respect to laser power (the value of laser power is measured at a constant location outside of the arcjet chamber). The LIF data approaches signal to noise limitations at lower laser powers and appears to deviate from the theoretical line of intensity proportional to power squared (a straight line when placed on a log-log plot) at higher laser powers. At low laser powers, the signal barely overcomes the background plume emission, and at higher laser powers it appears that processes such as saturation or ionization occur that prevent the measured fluorescence from continuing to grow at the same rate when laser power is increased beyond a certain point (in this example, about 50 microJoules of laser energy). It stands to reason that if the data desired can be taken at laser powers higher than the signal to noise limits, but still in the range where the fluorescence is proportional to power squared that the measured fluorescence will be representative of the emitting species present and the contributions to saturation, ionization and amplified spontaneous emission will be negligible or non-existenL 5 2 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. L IF t 0 2 Z 1 2 0.1 A SE e 0.01 50 70 90 10 30 Laser Power #J) Figure 4-2. Arcjet study of LIF signal dependence on laser power compared to ASE power dependence, which has a much sharper response to laser power. As a check of one of these processes. Figure 4-2 also shows returning laser intensity data measured along the line of the original input laser beam. The beam sent into the plume is at wavelengths near 205 nm. The last turning mirror prior to entering the vacuum chamber is 95% reflective for wavelengths near 205nm, but is 96% transparent to wavelengths in the 600 nm range. Because of this, a photomultiplier tube with a H a notch filter can be placed behind this turning mirror, and amplified spontaneous emission occurring in the plume between the n=2 and n=3 levels of the hydrogen atom (at 656 nm) can be measured by this instrument. At the laser energy where the LIF data deviates from power squared, 656 nm light along the 205 nm irradiating beam is measured at levels almost two orders of magnitude less than the LIF signal. As the 205 nm laser power is increased, the energy measured in amplified spontaneous emission increases at a rate 53 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. greater than input energy to the fifth and is as large as the smallest LIF signal by the time the beam is measimed to be 100 microJoules. At this laser energy, one can clearly see that the LDF signal is no longer proportional to the emitting species and is less in energy by at least the amount that would have been collected perpendicular to the laser beam. The laser energy appears to be instead pumping the ASE process back along the laser beam with energy possibly going into other processes as well. 4.1.2.1 Losses in Fluorescence Due to Amplified Spontaneous Emission LIF e 0) i I o 3 ta i A S E 205.135 2 0 5 .1 4 205.145 205.15 Laser Wavelength (nm) Figure 4-3. Narrower ASE spectrum compared with LIF spectrum. Figure 4-3 shows the differences in the lineshape between the measured ASE signal and the measured LIF signal in wavelength space when laser power was at the maximum level tested. Due to the differences in shape, compensating the LEF signal by the magnitude of the ASE loss mechanism through direct measurement of the ASE was deemed unreliable. As mentioned earlier, 54 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. an effort was made to collect data at laser powers where ASE loss was insignificant and the LtF signal did not deviate from being proportional to laser power squared. The success of this approach allowed data acquisition to be accomplished without the requirement of measuring the ASE during each wavelength scan. Instead, a sampling of integrated fluorescence signals at several laser powers (as demonstrated in Figure 4-2) was made prior to taking each set of data and was conducted up to several times per day as configurations were altered. Each time this sampling was done, verification of fluorescence proportional to power squared at the desired laser intensity was made prior to taking any further data. The temperature and number density data presented here are all obtained with a laser energy of about 0.03 mJ per pulse as measured at a constant location on the optical bench. Laser powers lower than this were found to approach signal to noise limits for the configuration, most notably when probing away from the center of the arcjet nozzle. Laser powers higher than this caused deviation from the power squared proportionality in both the arcjet plume and in the density calibration cell. The temperature/velocity calibration cell received a much smaller proportion of the laser energy than in the data collection volume in the arcjet or density calibration cell (about 5% of the total) and deviation of fluorescence from proportionality to laser power squared was never observed in the cell. 4.1.2.2 Temperature Prediction Effects When Fluorescence is Not Proportional to Laser Power Squared In Figure 4-2 the integrated fluorescence was observed to deviate from the laser power intensity squared, which demonstrated the saturation and ASE mechanisms that would directly impact the interpretation of the total fluorescence and hence the density of the probed species. 55 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Additionally, the deviation from the proportionality expected for a two photon process can lead to apparent broadening of the fluorescence lineshape which causes a misinterpretation of temperature. This broadening occurs as the center of the transition saturates and stops increasing while the wings of the transition continue to grow as laser power is increased. An example of how temperature can be misinterpreted as this “saturation broadening” takes place is shown in Figure 4- 4. 3500 3000 2 2500 i 2000 1500 5- 1000 500 20 60 100 140 180 220 Laser Power (jij) Figure 4-4. Apparent temperature for the arcjet at changing laser powers At laser powers higher than 50-60 microJoules (as measiu-ed at a constant location on the optical bench) the temperature measured (using the assumption that the width of the line is solely the Doppler width) appears to dramatically increase from the baseline of 1600-1700 K to temperatures twice as large. As the laser power falls below 50 microJoules or so, the temperature measured remains constant as saturation effects become insignificant or are not present. The temperature measurements for this work were made at 30 microJoules. 56 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 .1 .3 L if e t im e m e a s u r e m e n t s t o d e t e r m i n e t h e e f f e c t s o f C o l u s i o n a l Q u e n c h in g To use the fluorescence intensity data in Figure 4-1 for a relative density measurement, the integrated area of the fluorescence at one location is compared with the integrated fluorescence from another spatial location. To make an appropriate comparison, quenching effects due to the density of collision partners must be taken into account. Since it is expected that the density of collision partners is not constant throughout the plume, a correction for this collisional quenching must be made at every spatial location where data is taken. To accomplish this task, fluorescence lifetime measurements are made at all data locations by recording the fluorescence time decay on a fast digitizing oscilloscope. — Laser Pulse (-5 ns) - T = 0 nun (7 ns) ■ —r = 4.5 nun (13 ns lifetime) S t 5 10 15 0 20 25 30 35 40 Time (ns) Figure 4-5: Fluorescence decay for n=3 hydrogen atoms at two positions along the arcjet nozzle exit plane. For comparison, a trace of the laser pulse alone (with no hydrogen atoms present) is shown. 57 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Figure 4-5 shows temporal fluorescence of the laser beam alone, the fluorescence decay at the center of the plume and the fluorescence near the outer edge of the arcjet nozzle. The lifetime of the unquenched hydrogen 3-2 transition has been shown to be 15.7 ± 1.5 ns^' when irradiated with a laser similar to the one used for this experiment (the laser type may possibly affect the pumping ratios of the fine structure and alter the decay time). Some coUisional quenching is still evident even near the edge of the visible plume near the nozzle exit where the 1/e decay time (when the measured fluorescence decay is modeled with a best fit-least squares exponential decay as described in section 2.2.2.2) is approximately 13 ns. At the center of the plume near the nozzle exit, collisions are occurring rapidly enough that the typical fluorescence decay is about 7 ns. The measured fluorescence for the quenching observed at each of the spatial locations where data is taken can then be proportionally corrected, and a relationship between the different integrated signals can be made. The quenching correction profiles are first constructed, which, when combined with the relative fluorescence intensity profiles, lead to relative density profiles. With calibration of the fluorescence, the relative density profiles can then be converted to absolute density profiles. Figure 4-6 shows results for lifetime measurements from several separate days as a function of radial position from plume centerline near the arcjet nozzle exit. The lifetimes measured in the arcjet plume indicate coUisional quenching which could be due to electrons, H:, or most likely some combination of both. The quenching appears to be greatest in the center of the plume. Each data point is determined from analyzing the fluorescence decay at each spatial location (as in Figure 4- 5) and fitting the decay to an exponential. The 1/e time can be determined and used to calculate to the lifetime of the transition. The lifetime is then plotted against spatial location. 58 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2 11 = 1 0 - t ■ 6 ■ 2 0 2 6 4 Radial Position (mm) Figure 4-6: Fluorescence lifetimes as a function of radial position along the arcjet nozzle exit (0.4 mm away) from many different days. The line represents a best fit to the data. 15 14 13 12 11 10 9 8 7 0 2 4 8 10 12 6 16 18 14 Axial Position (nun) Figure 4-7: Fluorescence lifetimes as a function of axial distance from the arcjet nozzle exit along the plume centerline from three different days. The line represents a best fit to the data. 5 9 R epro(juced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Figure 4-7 shows a profile of quenching lifetimes axially from the arcjet nozzle exit along the plume centerline. An immediate increase in lifetime with distance, due to a decrease in collision partners as the plume expands, is observed with a lower gradient slope in the profile as the measurements are taken farther out in the plume. i 17 IS 13 11 9 7 5 -10 -8 ■ 2 8 -6 -4 0 2 10 4 6 Radial Position (mm) Figure 4-8: Fluorescence lifetimes as a function of radial distance taken 10 mm downstream from the arcjet nozzle exit on three different days. The line represents a best nt to the data. Figure 4-8 shows the lifetime profile for a radial distribution at a location 10 mm downstream from the arcjet nozzle exit. The asymmetry of the distribution is unexpected, and is somewhat repeatable (the data shown are from three different days), but it’s shape may be explained by potential facility interaction with the arcjet plume where collisions are more frequent at the edges of the visible plume. Additionally, the significance of each of the collision partners to the atomic hydrogen at this point in the expansion may be different than at the nozzle exit and collisions at this location may be dominated by a different species than the one which dominates at the arcjet nozzle exit. No effort was made in this work to determine the density profiles of molecular hydrogen, 60 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. hydrogen ions, or electrons at 10 mm downstream and at the arcjet nozzle exit. Regardless of the reason for the quenching, the correction for it is required to compensate for fluorescence lost due to collisions at the edges of the radial distribution. 4.1.4 DETERMINING A TOMIC HYDROGEN CONCENTRA TION IN THE CaUBRA TION CELL As discussed in section 2.3, NO: reacts rapidly with hydrogen atoms when the two gasses are completely mixed in the large drift tube volume. The calibration data shown in the following example involved a steady flow of 1 slpm of a 2% hydrogen in helium mixture through the density calibration cell described in section 2.3. A mixture of 2% NO: in helium is then added until the signal decreased, typically at about .08 slpm of NOVHe. A small vacuum pump brought the cell pressure down to 950 Pa (7.2 torr) during gas flow. à 1.5 0.5 0 4 6 8 1 0 1 2 NOj Partial Pressure (m Torr) Figure 4-9. Determination of atomic hydrogen density and emitted signal through titration of NO: into calibration cell. 61 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The LIF signal decreases until it is gone at the point where the partial pressure of added NO; is equal to the partial pressure of hydrogen atoms in the cell. Control of the amount of NO; added, allows the determination of the amount of hydrogen atoms present in the calibration cell. Figure 4- 9 shows data from one calibration run. The NO; concentration at the x intercept is equal to the concentration of atomic hydrogen present as each hydrogen atom is reacted away for each NO; molecule added as described earlier in section 2.3. Data is taken by increasing and decreasing the addition of NO; both progressively and randomly to insure reproducibility and eliminate time dependent effects. A single data point is determined by the average intensity of the center of the fluorescence line over 2 to 5 seconds of data acquisition. The amount of NO; added to the steady state flow is then altered and the pressure is given time to stabilize, usually taking about 20-30 seconds. The intensity, flow rates, and cell pressure are all measured in order to calculate accurate partial pressures. Overall pressure changes less than 5% during this measurement. From these measurements, determination of hydrogen fluorescence intensity in the calibration cell environment was related to actual atomic hydrogen density. With this information and the appropriate correction factors for quenching, arcjet measurements could be calibrated for absolute atomic densities. 62 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 .2 G r o u n d S t a t e A t o m ic H y d r o g e n M e a s u r e m e n ts a t 1.34 kW O p e r a t io n 4 .2 .1 O v e r v ie w All thruster data were taken on a 1-kW-class arcjet designed by NASA Lewis Research Center’' with cathode gap set to 1.78 mm (0.070”). The results shown in this section were all measured in the plume of an arcjet operating at 1.34 kW (134V, 10A) to allow comparisons (shown in later sections) with previous data taken at other research institutions. Section 4.3 will report data taken at a lower operating condition of 800 watts arcjet power. The arcjet operated on a hydrogen gas propellant flowing at 13.1 mg/s (8.74 slpm) into a chamber vacuum pressure of 6 Pa (45 mtorr). The gas flow and power were chosen to closely match the conditions under which the most complete set of previous diagnostic data for a 1 kW hydrogen arcjet were taken.'" Additionally, this facilitated comparisons between different data sets and between data and models.^^ Figure 4- 10 shows the V-I curve of the arcjet used, compared with the Stanford thruster for the same hydrogen flow. The figure indicates that a very close match has been achieved between the two experimental systems. Matching the V-1 curve can be accomplished by either adjusting propellant flow rate or cathode spacing with respect to the anode. The power supplies for these thrusters provide constant current and a larger cathode spacing can lead to higher potentials between electrodes for a given current. Additionally, higher propellant flow rates can elongate the arc also leading to higher voltages. The hydrogen flow controller for the arcjet as well as the flow meters for the discharge 63 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. cell were calibrated prior to test using a wet test meter to insure that the propellant flow rate matched previous reported flow rates. The cathode spacing was then set in order to match the reported V-I curve. 170 165 160 155 I' -1150 145 140 135 130 * Phillips Lab: Pobst, Wysong, et al • Stanford L'niversitv: Liebesklnd • ♦ J L ♦ • • ♦ ♦ # # ♦ ♦ . ■ I I . I 6 7 8 C urrent (Amps) 10 11 Figure 4-10: V-I Curve for 1 kW hydrogen arcjet operating with 13.1 mg/s of hydrogen. For comparison, the V-I curve with the same hydrogen flow is shown for the Stanford apparatus. The data is based on specific energy data published by Olin Aerospace regarding the Stanford measurements.^^ Data for this effort were taken in the arcjet plume in order to provide additional inputs to arcjet modeling efforts, both at the nozzle exit and in the outer plume region. Accurate modeling of the arcjet is expected to lead to more efficient thruster designs, while accurate modeling of the thruster plume is expected to play a significant role in satellite integration of these thruster systems, predicting where plume particles may go and what type of contamination to the spacecraft may or may not occur. 6 4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The data shown in the following graphs was all taken in the two dimensional plane depicted in Figure 4-11, though data has been taken in three dimensions to assure symmetry in the plume. The circles representing the control volumes are a much larger symbol size than actual physical size and are enlarged for the purposes of adequate viewing. Plume A rc je t iNo/y.Ic 0 ID O o o o o o o o o o o 25 30 Exit Plane dlftance (mm) Figure 4-11: A visualization of the physical location where data was taken in the arcjet pliune. The actual control volume of the fluoresced/detected area was measured to be 0.040 mm ± 0.01 mm. This measurement was accomplished by both determining beam waist size and matching the collection aperture to the size of the beam waist. The beam was measured by placing a razor blade attached to a motion control system at the location of beam focus with a ultraviolet photodiode in the beam’s path behind the razor blade. The beam is attenuated to protect the photodiode and the razor blade is placed so as to completely block the beam. The blade is moved out of the way such that the shape of the laser intensity is measured. Assuming the beam is Gaussian in shape, allows a determination of beam diameter based on the accuracy of the motion control step size and the bandwidth of the signal intensity. Measurements were made at several 65 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. different intensities to determine if the attenuation altered the beam waist. The size of the beam appeared to be independent of attenuation, though the range of powers that could be tested was limited due to the sensitivity of the photodiode. The optical collection diameter was adjusted by varying the size of the fixed diameter aperture and adjusting its location with respect to the focus lens that was set to collect light from the location of the beam. Verification was accomplished by placing a target that fluoresced due to the ultraviolet laser light at the control volume in view o f the collection optics. 4.2.2 N e a r NOZZLE EXIT The following data was taken near the nozzle exit about 0.4 mm downstream from the location determined to be the axial “zero" point. This "zero" point was determined to be the location where the LEF signal was attenuated by a factor of two due to the position of the nozzle body encroaching upon the optical collection volume as it was positioned using a fine motion control system. Since the arcjet nozzle would expand as it increased in temperature, the position of zero had to be monitored. Measurements would only commence after some stabilization of the nozzle expansion was observed. Periodically, the location of the nozzle exit would be verified to insure that full expansion had occurred for the operating arcjet power level where data was being taken. During the initial warming up period of the arcjet, which took 30 to 60 minutes depending upon the operating power, approximately 0.5 mm of expansion could be observed in the 1 kW arcjet nozzle . 6 6 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 .2 .2 .1 A x ia l V elo city Figure 4-12 shows the radial distribution of axial velocities 0.4 mm from the arcjet nozzle exit. The velocities were measured with a laser beam entering the plume of the arcjet axially. Data for the plot were taken over several days and the scatter may indicate day to day operational changes in the arcjet as well as measurement and alignment uncertainties. Data taken in one sitting was continuous spatially, and changed in a very smooth manner. 14 ### 12 10 8 # » 6 4 2 0 4 -2 0 2 6 4 -6 Radial Position (mm) Figure 4-12. Profile across nozzle exit plane (0.4 mm from exit) of the axial velocity component of the groimd state hydrogen atoms containing data from several different days. 67 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. A peak velocity of 13 km/s to 14 km/s is observed at the centerline of the nozzle exit plume with axial velocities dropping to near 6 km/s at 4 mm radially away from the center. Due to significant background emission when imaging through the plume, measurements of velocities past 4 mm radius were only possible on the side of the plume nearest the optical detection. Velocities down to below that measurable by this LIF technique were observed between 4 mm and 6 mm radially away from the centerline. It should be noted that velocity profiles of the ground state hydrogen atoms are in close agreement with previous measurements using excited state LIF.'‘ ‘ A comparison of this data with these previous excited state measurements is made in section 4.4.1.1. The profiles are extremely sensitive to spatial centering of the arcjet nozzle with respect to the optical collection volume in both vertical and horizontal directions and an alignment procedure is performed prior to evacuation of the chamber. After arcjet operation commences and a period of heating (approximately 30 minutes to an hour) takes place, sample velocity distributions are taken horizontally and vertically to center the optical detection with the peak velocity of the arcjet plume. This could require an adjustment of 0.25 mm to 0.5 mm from the initial alignment at atmosphere in one or both of the axes that make up the nozzle exit plane. 4.2.2.2 Radial Velocity Figure 4-13 shows the corresponding distribution of the radial component of velocity when the laser beam is brought to the plume in a line parallel to the nozzle exit. Observing the Doppler shift in the radial direction allows the component of velocity perpendicular to the bulk flow to be measured by the beam. 6 8 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 6 ■ 4 4 0 6 Radial Position (mm) Figure 4-13. Profile across nozzle exit plane (0.4 mm from exit) of the radial velocity component of the ground state hydrogen atoms containing data from several different days where the radial position is measured parallel to the incoming laser beam. To measure the data shown in the figure, the arcjet body is translated such that the intersection of the optical collection focus and the laser beam probes locations from one edge of the exit plane plume to the other, in the direction that the laser beam is crossing the nozzle exit. When the flow is moving away from the laser beam (i.e. when the radial component of flow from the side of the nozzle exit opposite of where the laser beam originates is being measiued) velocities wiU have negative values. Velocities will have positive values when the flow is moving toward the laser beam (i.e. flow on the side of the nozzle exit near the origination of the laser beam). Near zero radial velocity is measured in the center of the nozzle exit as would be expected. In this particular 6 9 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. case, positive position values for the arcjet body are locations where the plume being measured is near the location where the beam is brought radially across the thruster. Radial velocities from zero in the center to 6 km/s at the edges of the measurable plume (5 mm radius) were observed increasing from the center in a very linear fashion. 4.2.2.3 Azimuthal Velocity 4 3 2 1 0 -I -2 -3 -4 -3 -2 2 3 4 -4 -I 0 1 Radial Position (mm) Figure 4-14. Profile across nozzle exit plane (0.4 nun from exit) of the azimuthal velocity com ponent (indicating swirl) of the ground state hydrogen atoms containing data from several different days where the radial position is measured perpendicular to the incoming laser beam . Azimuthal velocity is measured in much the same way as radial velocity is measured with the movement of the arcjet body (while the optics remain stationary) being the primary difference. The measurement of radial velocity takes place along an axis of points at nozzle exit that is 70 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. perpendicular to the bulk flow and is also perpendicular to the axis of the incoming laser beam. In this way, the azimuthal component of velocity or “swirl" can be measured. As is seen in Figure 4-14, the amount of swirl seen in the arcjet plume at nozzle exit is minimal and centers about 0 km/s with measurements both positive and negative on either side of the thruster. No significant evidence of swirl in ground state atomic hydrogen velocities is observed, 4 .2 .2 .4 T em p era tu re 2000 1500 I 1000 a. 500 Radial Position (mm) Figure 4-15: Temperature profile across the nozzle exit plane (0.4 mm from exit) of the ground state hydrogen atoms. Graph contains data from several different days. Figure 4-15 shows a temperature profile of data points from many different days taken along the nozzle exit. The results indicate a peak temperature of 1600 K - 1800 K in the center, 71 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. dropping as much as 1000 K over the 5 ram radius of the nozzle exit. The low signal to noise of the fluorescence (when low laser power must be used, as was described above) makes measurement of the Doppler width difficult, especially at the edges of the plume. Doppler shift measurements (used for velocity measurements) are less affected due to the correlation of many different data points when determining the location of the Gaussian peak. Temperature, based upon the square of the measured Doppler width, has, therefore, substantially greater uncertainty associated with it than does velocity for this type of optical measurement. The uncertainty due to scatter can be seen in the plot, and ranges from. ±200 K in the center to almost ±400 K at the edge of the plume. Because the temperature is based upon the determination that the width of the spectral line measured in the plume is due solely to the Doppler broadening effect, any errors in temperature prediction would be ignoring a broadening effect. Actual temperatures, if different because of an error in that assumption, would be lower rather than higher. Details of the validity of the assumption are given in sections 2.2.2 and 2.2.3 4.2.2.S Density As described earlier, the total integrated measured fluorescence from the collection volume in the plume can be compared to a calibration measurement in order to determine absolute number density of the ground state hydrogen atoms. To correlate the two fluOTescence measurements, one in the plume and one in the calibration cell, corrections for quenching differences due to the different ambient densities must be taken into account. 72 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.2.2.5.1 Quenching Correction o f Fluorescence for Density Measurements The need for correcting for coUisional quenching is evident in Figure 4-16, where the density prediction before, and after, the correction for quenching is seen. In this case the significance of the correction is found to be large in the center region where coUisional densities are high in the plume, and minimal at the edge of the profUe where coUisional interactions are minimal as in the calibration ceU. 1x10 density corrected for quenching 9x10 8x10 7x10 m i 6x10 density not corrected for quenching ^ 5x10 a I 4xlo' S 3 I ^ 3x10 2x10 1x10 -5 3 -3 -2 2 4 -4 -1 0 5 1 Radial Position (nun) Figure 4-16. Significance in correcting for quenching when determining density from a relative fluorescence profile. Note that the largest correction occtn-s in the center of the profile where the quencfiing due to collisions with the hydrogen molecules or electrons is greatest and is minimal at the edges where quenching is relatively insignificant. 73 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The correction values are determined from lifetime measurements in the plume and in the calibration cell. The lifetime measurements taken for the radial profile shown in Figure 4-16 are described in more detail in section 4.1.3. 4.2.2.5.2 Relative Density Profile Upon correcting the relative fluorescence data for quenching, data from one day to another may be compared to each other as relative density data. Figure 4-17 shows relative density profiles from several different days. Agreement between the different data sets is extremely good and all exhibit a centerline peak with a decrease in density down to 10% of centerline density over the course of 3 to 4 mm. 1.00 - 1 a b £ E 3 « 0.10 I 0.01 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 Radial Position (mm) Figure 4-17: Profile across nozzle exit plane (0.4 mm from exit) of the relative number density of ground state hydrogen atoms containing data from several different days. 74 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Data points in Figure 4-17 are connected by lines to indicate the continuous nature of the points that were taken on the same day. Data taken in one session had smooth transitions from point to point and was found to be quite repeatable during a single session. 4.2.2.S.3 Single Day Calibrated Density Profile Absolute number density data is constructed from measured relative fluœescence intensity data. The fluorescence intensity data is taken at several locations in the plume only after determination of the appropriate laser power setting. Laser power settings are determined based upon a daily laser power vs. integrated fluorescence test to insure that processes such as saturation, amplified spontaneous emission and multiphoton ionization are not present in the plume or calibration cell environment at the specific laser intensities used for the fluorescence. The measured plume fluorescence data is then corrected for quenching reductions in intensity based upon a daily measurement of transition lifetime vs. spatial position taken at each point in the plume where density data is measured. The resulting relative density data is then calibrated based upon a daily measured fluorescence vs. varying atomic hydrogen concentration test performed in an in-sini calibration cell at the same laser power and with the same optical senip as the plume fluorescence data. Some days the calibration occurs prior to plume measurements, other days it may occur after plume measurements. The resulting calibration converts the relative density data to an absolute number density profile across the spatial range where the data is taken. Absolute number density data from a single day’s operation is shown in Figure 4-18. 75 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.2.2.5.4 Multiple Day Calibrated Density Profile Density data was taken on several different days in the manner previously described in section 4.2.2.5.3. Density calibration data was taken on each day that density data was measured, and the calibration data taken for one set of measurements is only used to calibrate those particular measurements. It was observed that the day-to-day changes in optical setup, though extremely slight, had a significant impact on the measurements due to the sensitivity of the measurement to the combination of daily variations in laser power, lens placement, beam focus size, thruster alignment, collection optics, etc. and, thus, needed to be accounted for each data collection session. c w Ç a u 3 z 16 1x10 15 1x10 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 Radial Position (mm) Figure 4-18: Profile across nozzle exit plane (0.4 mm from exit) of the absolute number density of ground state hydrogen atoms measured on a single day. 76 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Figure 4-19 shows the calibrated number density profile of hydrogen atoms near the exit plane of the nozzle (0.4 mm from nozzle exit plane). Separately calibrated data sets from many different days are included in the plot. The peak of the density profile is at the center of the nozzle exit plane and indicates a number density of order 1 x 10‘® cm^ which falls off about one and a half orders of magnitude over the radius o f 5 mm out to the nozzle exit wall. É 1 a I 16 1x10 IS 1x10 • • 14 1x10 -2 2 4 6 -6 -4 0 Radial Position (mm) Figure 4-19. Profile across nozzle exit plane (0.4 nun from exit) of the absolute number density of ground state hydrogen atoms containing data from several different days. The significance of the magnitude is discussed in more detail in sections 4.4 and 4.5. The magnitude of the number densities shown above, were somewhat surprising in that they were significantly higher than had been predicted by several computational models. Comparisons 77 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. between the velocity, temperature, and density data and the predictions from several arcjet models are given in section 4.4. 4.2.3 IN THE ARCJET PLUME 10 MM FROM NOZZLE EXIT Downstream in the arcjet plume, another measurement “cut” was made parallel to the nozzle exit plane in order to observe the expansion of the plume and the properties of the atomic hydrogen species in the expansion (see Figure 4-11). Comparing this data with downstream plume models is also expected to aid modelers in their predictions of surface contamination due to the arcjet plume. 4.2.3.1 Axial Velocity I > > ë > 14 12 10 8 6 4 2 0 -10 -5 0 5 10 Radial Position (mm) Figure 4-20: Radial profile of ground state hydrogen atom axial velocity across the arcjet plume, 10 nun downstream of (and parallel to) the nozzle exit plane. Plot contains data from multiple days. 78 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Figure 4-20 shows the atomic hydrogen velocity profile 10 mm away from the arcjet nozzle exit. Note that the peak velocity along the plume centerline remains near 13 to 14 km/s as in the nozzle exit velocity data in Figure 4-12, while the wings are much broader and velocities are as high as 7 to 8 km/s at 10 mm away from the centerline. This is consistent with expansion of the propellant into the vacuum and is compared to plume expansion computations in section 4.4. 4.2.3.2 Radial Velocity I > I 8 6 4 2 0 1 -4 -6 8 -6 8 8 -4 2 0 7 6 10 -10 4 Radial Position (mm) Figure 4-21: Radial profile of ground state hydrogen atom radial velocity across the arcjet plume, 10 mm downstream of (and parallel to) the nozzle exit plane. Plot contains data from multiple days. Ten millimeters downstream of the nozzle exit, radial velocity is measured in the same manner as was done near nozzle exit in section 4.2.2.2. Again, the radial velocity ranges from 0 to 6 km/s with Figure 4-21 showing positive velocities when the gas flow was moving toward the probing laser beam and negative velocities when the gas was moving away from the beam. The magnitude 79 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. of the maximum cross-velocity vectors does not appear to be diminished downstream with respect to the nozzle exit measurements, indicating an expansion of atoms that is not inhibited by chamber background pressiu’ e. Measurements past 10 nun in plume diameter were not possible at this location as the species became too diffuse to supply enough emitters for detectable fluorescence. 4.1.3.3 Temperature 3 2 u a I 2000 1500 1000 500 0 -10 0 5 •5 10 Radial Position (mm) Figure 4-22: Radial temperature profile of ground state hydrogen atoms across the arcjet plume, 10 mm downstream of (and parallel to) the nozzle exit plane. Plot contains data from multiple days. Figiue 4-22 shows measured temperatures 10 mm downstream of the nozzle exit ranging from 0 to 10 mm away from the axial nozzle centerline. As discussed in earlier sections, the scatter in 80 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. temperature measurements is larger than is observed in either velocity or density measurements due to the difficulty in measuring the fluorescence linewidth, and the resulting width-squared analysis to derive temperature. Note that the temperature values are significantly lower than those measured at nozzle exit in Figure 4-22 (the plots are shown using identical vertical axes for the purposes of comparison). Whereas the temperature profile at nozzle exit is peaked with a maximum measured to be near 2000 K, the profile 10 mm downstream is cooler (maximum temperatures near 1000 K) and less defined, with any profile structure being hidden within the scatter of the measured values. 4.1.3.4 Density 1x10 I I 1x10 U E 3 Z 1x10 ■ 5 0 -10 10 3 Radial Position (mm) Figure 4-23: Radial profile of ground state hydrogen atom density across the arcjet plume, 10 mm downstream of (and parallel to) the nozzle exit plane. Plot contains data from multiple days. 81 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Ground state atomic hydrogen absolute density data was measured at a location in the plume 10 mm downstream of the arcjet nozzle exit. Densities were found to be significantly lower at this location than at nozzle exit dropping from a peak density of -1 x 10‘® cm ' to one -2.5 x l0 ‘^ cm \ The decrease in magnitude and widening of the density distribution is consistent with plume expansion expectations. Figure 4-23 shows the density profile measured from 0 to 10 mm away from the axial centerline of the plume. Note that the density magnitude axis is the same as that used to depict the nozzle exit data in Figure 4-19. 4.2.4 ALONG THE CENTERUNE AXIS FROM NOZZLE EXIT TO 30 MM DOWNSTREAM The expansion of the plume is also observed tfirough measurements of plume properties along the nozzle centerline from nozzle exit to 30 mm downstream. Measurement of the plume data along an axial line allows observation of the plume expansion into vacuum of a single plume property along a line of gas flow analogous to a classical streamline. Measurement locations are shown in Figure 4-11. 4.2.4.1 A xial Velocity Figure 4-24 shows axial velocity data taken along the plume centerline from the arcjet nozzle exit to 28 mm downstream. Note that the velocity which is between 12 and 14 km/s peak at the arcjet nozzle exit essentially remains unchanged along the centerline demonstrating that the interactions with the ambient chamber background pressure are minimal. 8 2 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. u > 14 12 10 8 6 4 0 5 0 10 15 20 25 30 Radiai Position (nun) Figure 4-24: Axial profile of ground state hydrogen atom velocity along the arcjet plume centerline. Plot contains data from several different days. 4.2.4.2 Temperature Atomic hydrogen ground state temperature measurements along the plume centerline are shown in Figure 4-25. Species temperature appears to decrease from the temperature range of 2000 K to 2500 K near the nozzle exit to a range of 600 K to 12(X) K over the first 20 mm of plume expansion. Changes in temperature farther downstream in the plume are not discernible. This decrease in species temperature as one moves farther out into the plume indicates that the mechanisms which provide energy to the gas are less likely to be from re-absorption of arc photons as has been hypothesized in previous research,^ and are more likely resultant from energy transfer in the arcjet nozzle with a decrease in temperature from the ncrmal expansion-relaxation collisional cooling process. 83 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2500 2000 g 1500 2 a. I 1000 500 0 10 15 30 20 25 Axial Position (mm) Figure 4-25: Axial profile of ground state hydrogen atom temperature along the arcjet plume centerline. Plot contains data from several different days. 4.2.4.3 Density Decrease in atomic hydrogen number density along the plume centerline is shown in Figure 4- 26. A drop of almost two orders of magnitude in density is observed from the nozzle exit to 30 mm (over approximately 3 nozzle exit diameters) decreasing in a logarithmic fashion. Decrease in atomic density is expected to be comprised of number density reduction due to free jet expansion coupled with some recombination of atomic hydrogen species. 8 4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1x10 I I E 3 Z 1x10 1x10 0 20 25 5 15 30 10 Axial Position (mm) Figure 4-26: Axial profile of ground state hydrogen atom temperature along the arcjet plume centerline. Plot contains data from several different days. 4.2.5 IN THE N o z z l e E d g e Wa k e R e g io n 4.2.5.1 Density Measurements of atomic hydrogen density were conducted in the immediate wake region near the edge of the arcjet nozzle. The impetus for conducting such measurements was twofold. First, computational models using Direct Simulation Monte Carlo techniques can be calibrated for a particular simulation by using a low density region of the simulation that is sensitive to changes in the high density portion of the simulation. This allows determination of proper simulation of the flow patterns, and also indicates how well suited the number of simulated particles is for solving 85 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the problem that involves substantially more real particles (such as atoms, molecules, and electrons). The immediate wake region near the nozzle exit is a good calibration region for the arcjet plume as it exists “behind” the main plume region, having lower densities than the bulk of the plume. The wake number densities are based upon lower probability occurrences such as collisions in the expansion or upstream diffusion, each of which is sensitive to the density of the bulk plume region. Secondly, measurements taken near the nozzle wake region were expected to help determine if lower densities of atomic hydrogen than had been measured to this point could be observed. One reason that this was expected to be possible was that the background emission in the nozzle wake region was not expected to be as significant as in the main plume, thus possibly allowing the technique to be used on smaller number densities than were possible at the edges of the measurable plume where background emission prevented observation of the fluorescence of number densities below -1 X 10'"'cm \ Although, these measurements were never compared to a computational simulation, wake entrainment data was taken and is shown in Figure 4-27. Open circles in the figure display the two dimensional locations of a measurement, and the color at that location indicates density magnitude using the scale provided. The color gradients between each point were automatically determined by the plotting software chosen (Stanford Graphics 3.0d) using an inverse distance weighting between data points (the data is forced to zero density at the nozzle wall) and the gradients are displayed to show some semblance of continuity between the points. The color mapping, though, is in no way a substantive prediction of density gradients as, for example, higher scatter in the data at the very low density region of 10‘‘ cm ’ causes color patterns between data points that should likely be disregarded. 86 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. s s O es O C 5 r 5.5 6 6.5 7 7.5 8 8.5 9 I ' ' ' ' I ' ' I ' I ' ' ' ' I ' ' ' ' I - 1.5 -1 - 0.5 0 0.5 Axial Position (mm) 'X C -T ' ' - Number Density (cnr^) 3.0E + 13+ 2.7E + 13 to 3 2.4E + 13 to 2. 2.1E + 13 to 2, 1.8E + 13 to 2, 1.5E + 13 to 1. 1.2E + 13 to 1, 9.0E + 12 to I, 6.0E + 12 to 9, 3.0E + 12 to 6. experim ent OE+13 7E + 13 4E + I3 lE + 13 8E + 13 SE+13 2E + 13 0E + I2 OE + 12 Figure 4-27: Arcjet Nozzle Wake Region Density map 87 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 .3 G r o u n d S t a t e A to m ic H y d r o g e n M e a su r e m e n t s a t 8 0 0 W O pe r a t io n 4.3.1 Overview As in section 4.2. all data in the following section were taken on a 1-kW-class arcjet designed and provided by NASA Lewis Research Center.^' Cathode gap was set to 1.78 mm (0.070”) and the electrical properties of the arcjet saw a potential o f 154V between cathode and anode, when a fixed current of 5.2A was applied. This 800 watt operation was chosen to provide experimental data on thruster operation at a different power level than the 1.34 kW case depicted earlier and allowed for comparison with additional molecular density data taken at another research institution at the same power level.‘‘ For computational modelers, taking identical measurements at multiple power levels allows observance of the how the same geometric simulation compares to experimental data when the current through the arc is reduced, and the length of the arc (or the potential) is somewhat increased. By providing data at two power levels, it is hoped that the experimental information better addresses comparisons of approaches in modeling the electrical transport properties than it otherwise might if data were only taken at one power level. As in section 4.2, the data shown in the following graphs were all taken in the two dimensional plane depicted in Figure 4-11, though some data has been taken in three dimensions to assure symmetry in the plume. The circles representing the control volumes are a much larger symbol size than actual physical size and are enlarged for the purposes of adequate viewing. 8 8 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The actual control volume of the fluoresced/detected area was measured to be 0.040 mm ± 0.01 mm and is discussed in greater detail in section 4.2. 4.3.2 N e a r NOZZLE EXIT The following data was taken near the nozzle exit about 0.4 mm downstream from the location determined to be the axial “zero” point. This “zero" point was determined to be the location where the LIF signal was anenuated by a factor of two due to the position of the nozzle body encroaching upon the optical collection volume as it was positioned using a fine motion control system. Since the arcjet nozzle would expand as it increased in temperatiu’ e, the position of zero had to be monitored. Measurements would only commence after some stabilization of the nozzle expansion was observed. Periodically, the location o f the nozzle exit would be verified to insure that full expansion had occurred for the operating arcjet power level where data was being taken. During the initial warming up period of the arcjet, which took 30 to 60 minutes depending upon the operating power, approximately 0.5 mm of expansion could be observed in the 1 kW arcjet nozzle. As detailed descriptions of the method of data acquisition have been previously given in section 4.2, the following sections will by default refer to the corresponding 1.34 kW measurement subsection discussed in section 4.2 and will not continually repeat common details to your annoyance. In general, the technique in acquiring the data was identical for each measurement and differences (if any) will be discussed where appropriate. The following 800 watt measurement sections may then appear more terse than their previous counterparts in the 1.34 kW section. 89 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 .3 .2 .1 A x ia l V elo city I • • 0 -6 -4 -2 2 4 6 Radial Position (mm) Figure 4-28: Profile across the nozzle exit plane (0.4 mm from exit) of the axial velocity component of the ground state hydrogen atoms containing data from several different days. Axial velocity was measured radially across the nozzle exit plane (0.4 nun away from the "zero" point) and a profile of this data is shown in Figure 4-28. Peak atomic hydrogen velocity is seen in the centerline of the plume near nozzle exit and is measured to be near 12 km/s. The velocity is observed to drop to velocities below those measurable by this technique over 6 mm in radius. Note that at 800 watt operation, the background emission was significantly less than at 1.34 kW, and velocity measurements down to almost zero velocity are observed on both the near and far sides with respect to the collection optics. In general terms, the signal-to-noise was significantly better at 800 watt operation, though it did vary by location and measurement as at 1.34 kW operation. If like measurements are to be 90 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. performed, it is highly recommended that 800 watt arcjet operation be used rather than the 1.34 kW operation for initial proof-of-concept tests. Unfortunately, this set of measurements was conducted in the reverse order. 4.3.2.2 Temperature 2000 1500 S 1000 500 -4 4 ■ 6 -2 0 2 6 Radial Position (mm) Figure 4-29: Temperature profile across the nozzle exit plane (0.4 mm from exit) of the ground state hydrogen atoms. Graph contains data from several different days. Figure 4-29 shows the results from temperature measurements made in the arcjet plume near nozzle exit. As is first observed, the scatter in the data is large. A discussion on temperature data sensitivity is given in earlier sections (including section 4.2) and it is believed that temperatures on the lower end of a scatter group are more probable than the higher temperatures due to the propensity of error mechanisms that falsely increase temperatime rather than reduce it. 91 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Regardless of how the data points are weighted, temperatures in the center of the plume are observed to reach between 1000 K and 1500 K and reduce in magnitude radially in the plume to near 500 K at 6 mm radius. This shows significant cooling compared to the expected arc-gas temperatures of several thousands of degrees in the arcjet nozzle a few mm or less away. 4.3.2.3 Density 1x10 S a È s s 1x10 z 1x10 2 4 -6 -4 -2 0 6 Radial Position (mm) Figure 4-30: Profile across nozzle exit plane (0.4 mm from exit) of the absolute number density of ground state hydrogen atoms containing data from several different days. Ground state atomic hydrogen number density vs. radial position is shown in Figure 4-30. Peak number densities are observed in the center of the nozzle exit profile with magnitudes between 6x10'^ cm'^ and 1 x 10‘* cm’’ at the center and near 7x10" cm’’ at 4 nun from center. The preponderance of atomic hydrogen at 800 watt operation appears to be similar to that observed at 92 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1.34 kW operation. While the integration of the profile does show that less atoms are present at the lower operating power, the similarity of peak number density magnitudes is interesting. More detailed comparisons are given later in section 4.4.2. 4.3.3 IN THE ARCJET P l u m e 10 MM FROM NOZZLE EXIT 4.3.3.1 A xia l Velocity 14 12 10 8 6 4 2 0 .10 -5 0 5 10 Radial Position (nun) Figure 4-31: Radial profile of ground state hydrogen atom axial velocity across the arcjet plume, 10 mm downstream of (and parallel to) the nozzle exit plane Plot contains data from multiple days. Figiue 4-31 shows the atomic hydrogen velocity profile 10 mm away from the arcjet nozzle at 800 watts operation. Note that the peak velocity has dropped slightly from the peak velocity at the nozzle exit and is near 11 km/s rather than the 12 km/s observed at the nozzle exit. An axial view of velocity down the plume centerline will be shown in a later section. As observed with the other 93 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. radial velocity data taken, the species velocities are highest in the center of the plume and decrease at locations farther from the plume centerline. At 10 mm away from the plume centerline, the velocity at 10 mm downstream from the nozzle exit was observed to drop to between 5 km/s and 6 km/s. Past this radial point, measurements became difficult due to signal to noise levels. 4.3.3.2 Temperature 2000 1500 s 1000 500 -10 Radial Position (nun) Figure 4-32: Radial temperature profile of ground state hydrogen atoms across the arcjet plume, 10 mm downstream of (and parallel to) the nozzle exit plane. Plot contains data from multiple days. Figure 4-32 contains the atomic hydrogen ground state temperature data at 10 mm downstream of the arcjet nozzle exit. Thought the scatter in the data increases as the measurements are taken farther from the plume centerline (and therefore signal to noise decreases) the plume appears to have a rather flat profile perhaps decreasing as data is measured farther away 94 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. from the centerline, though the scatter prevents identification of any conclusive trends. In any case, temperatures averaging 800 K are significantly cooler than those measured near the arcjet nozzle exit. 4 3 .3 .3 D e n sity 1x10 ? I 1 O f = 1x10 S 1x10 -5 0 5 Radial Position (nun) Figure 4-33: Radial profile of ground state hydrogen atom density across the arcjet plume, 10 mm downstream of (and parallel to) the nozzle exit plane. Plot contains data from multiple days. Figure 4-33 presents the radial distribution of arcjet atomic hydrogen ground state density 10 mm downstream of the hydrogen arcjet operating at 800 watts power. The density scale of the plot is identical to the other radial density plots for enhanced comparison of the density magnitudes and profiles at the different positions and operating conditions. Note that over the 10 mm distance along the centerline, the peak density in the center has dropped almost one order of magnitude. At 95 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. this location in the plume radial drops of another order of magnitude from the peak are observed over 8-10 mm radially. 4.3.4 ALONG THE CENTERUNE AXJS FROM NOZZLE EXIT TO 30 MM DOWNSTREAM 4 .3 .4 .I A x ia l Velocity 1 "3 14 12 10 8 6 4 2 0 0 5 10 15 20 25 30 Axial Position (mm) Figure 4-34: Axial profile of ground state hydrogen atom velocity along the arcjet plume centerline. Plot contains data from multiple days. Figure 4-34 shows a profile of centerline atomic hydrogen velocity along the plume axis from the nozzle exit to 30 mm downstream in the plume. Velocity magnitude appears to stay level throughout out the plume after dropping a bit from 12 km/s to 11 km/s in the first 16 mm or so. The appearance of a change in velocity may be due to insufficient statistics in the measurement at the nozzle exit location and is not necessarily a significant or certain occurrence as this behavior is 96 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. not observed at the other power level and is not predicted by the computational models that will be shown later. 4.3A.2 Temperature 2500 2000 1500 # # E 1000 500 5 15 20 0 10 25 30 Axial Position (nun) Figure 4-35: Axial profile of ground state hydrogen atom temperature along the arcjet plume centerline. Plot contains data from several different days. Figure 4-35 shows the axial profile along the plume centerline of the ground state hydrogen temperature. Again, the vertical temperature scale is taken from the previous temperature plot in order to facilitate comparisons between the two operating power levels. Note that the temperature average at the nozzle exit of 13(X) K drops almost linearly with position until near 14 - 15 mm away from the nozzle exit. From this location out to the 30 mm point, the temperature is observed to remain constant within some level of scatter around 700 K. 9 7 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 .3 .4 .3 D en sity S w I a s 3 Z 16 1x10 IS 1x10 14 1x10 5 10 15 25 0 20 30 Axial Position (mm) Figure 4-36: Axial profile of ground state hydrogen atom temperature along the arcjet plume centerline. Plot contains data from several different days. Figure 4-36 depicts the decrease of atomic hydrogen number density along the plume’s centerline axis. Note that the appearance of decrease on the semi-log scale is rather linear indicating a logarithmic-like decrease in density over the first 25 to 30 mm away from the arcjet nozzle as was observed in Figure 4-26. 4 .4 C o m p a r is o n o f M e a s u r e m e n t s w i t h M o d e r n C o m p u t a t i o n a l P r e d i c t i o n s a n d O t h e r E x p e r i m e n t a l R e s u l t s One of the objectives of this experimental effort was to provide data that was of value to the electric propulsion community. With ongoing efforts in computational modeling of the hydrogen 98 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. arcjet and other experimental efforts also reporting data, an effort was made to create a community consensus on a “standard operating condition." When reporting results, experimentalists in the community would include results from this standard condition, and computational modelers would run their simulations at this operating condition as well. Because scientific researchers are often independent, common operating conditions or configurations are seldom observed across the work of many researchers. This effort took place under the auspices of the United States Air Force at the Phillips Laboratory’s Electric Propulsion Lab. A majority of the basic research on arcjets in the U.S. is funded and directed by the Air Force through the Air Force Office of Scientific Research. Due to this common thread between many arcjet researchers, computational results and experimental data began to be published and presented using the same common operating conditions that this work also adhered to. The following section presents results from other researcher’s work provided for use in comparison with the experimental data presented here. The different results are published and described in greater detail elsewhere as referenced in the following sectioiL The data from this experimental effort has correspondingly been provided to these other researchers for assistance in model calibration and other comparisons. 4.4.1 NOZZLE EXIT ARCJET MEASUREMENTS AT 1.34 KW OPERA TION The data in the following section is measured or simulated from the plume of a 1.34 kW hydrogen arcjet. Experimental data is presented from this work and from two other laser experiments conducted at Stanford University: excited state laser induced fluorescence'" and Raman Spectroscopy. Computational simulation results are provided from several sources. The results from the Boyd model are calculated by I. D. Boyd at Ccmell University.^’ The model is a 99 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Direct Simulation Monte Carlo (DSMC) particle code that predicts physical properties through most of the arcjet nozzle and out into the plume. The DSMC model does not include the entire arc process, however, and uses the Butler model inside the arcjet nozzle past the arc attachment point as a start condition. The Butler model is authored by G. W. Butler at Olin Aerospace Company (now Primex Aerospace Company) and is a single temperature Navier Stokes (NS) code that computes the flow properties throughout the arcjet nozzle, stopping at the nozzle exit.® A two temperature NS code written by Megli, Krier, and Burton from the University of Illinois predicts results that are also shown in the following figures.^® Like the Butler model the Illinois calculations extend to the nozzle exit, but not into the plume. Data from another two-temperature NS simulation (wtiich are also modeled the region just up to the nozzle exit) are provided by V. Subramaniam at Ohio State University. The model includes additional physical chemistry processes that are not modeled directly in the other NS codes. 4.4.1.1 Velocity Data and Models Figure 4-37 shows a radial profile of axial velocity at the nozzle exit of a 1.34 kW hydrogen arcjet with data from this work (previously shown in Figure 4-12) plus data from an excited state laser induced fluorescence measurement'"' which measured the axial velocity of atomic hydrogen species with electrons already in the n=2 state prior to irradiation. As is observed in the figure, the velocity of the ground state species matches well with the velocity of the excited state species indicating that a velocity slip condition between the states is not present. Additionally, the computational codes (both Navier-Stokes and DSMC) aU predict similar velocity profiles to each other with trends very similar to the experimental data, but slightly less in magnitude. Note that the Boyd DSMC model has a slightly different profile shape that does appear to follow the data 100 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. bener in the wings of the distribution. This is likely due to the fact that the DSMC code allows the simulation to go past the nozzle exit boundary and the DSMC simulation better represents the location where the data was taken 0.4 mm downstream of the nozzle exit. I 5 1 > 14 12 10 8 6 4 2 0 -6 o # # # • # Ground State H data (Pobst) O Cappelli Excited State LIF data ' Butler CFD Model KrIer CFD Model - • • - Boyd DSMC Model Subramanium CFD Model I I I I I I I I I I __ * . i -2 0 2 Radial Position (mm) Figure 4-37. Profile across nozzle exit plane (0.4 mm from exit) of the axial velocity component of the ground state hydrogen atoms containing data from several different days. For comparison, corresponding modeling and experimental data is 4.4.1.2 Nozzle Exit Temperature vs. excited state LIF, Raman, models Figure 4-38 shows the atomic ground state temperature measurements with additional experimental data provided from published excited state temperature m easurem ents,m olecular rotational temperatures,*'and computational models.®’ ^'^^'^* 101 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 3500 h 3000 2500 t 2000 s I 1500 1000 500 0 -6 o ^ # G round State H d ata (Pobst) O Cappelli Excited State L IF data '93 O Cappelli Excited State LIE data '96 □ Cappelli H2 Rotational R am an data ........ Butler CFD Model K rier CFD Model - - - - Boyd DSMC Model T t — ^braiMnlpm CFD Mod^l -2 0 2 Radial Position (nun) Figure 4-38: Temperature data (previously shown in Figure 4-15) with corresponding data (same arcjet conditions, same profile location) for LIF of the electronically excited hydrogen atoms, Raman molecular rotational temperatures,*^ and Computational m odels.*'^*'^’'^ * While the temperatures measured from the ground state hydrogen atoms are quite similar to the rotational molecular temperatures, temperature measurements of the ground state hydrogen atoms are about a factor of two cooler in the center than those of the excited-state hydrogen atoms as originally reported in 1993.'^ This led to the early assessment that the data indicated a significant non-equilibrium between the excited state and ground state hydrogen atoms. A 1996 revision of this data, taking account of additional Stark broadening corrections that are applicable to this excited state, reduced the magnitude of this data significantly, such that the corrected 1 0 2 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. temperatures appear to closely agree with the ground state temperature measurements.'^ While this does not disprove the presence of non-equilibrium between states, this measurement is no longer data that can be used to support this assumption. Observing the computational results, the single temperature model by Butler shows temperamre results that tend to follow the temperatures originally reported from the hydrogen excited states. The two temperature Megli model also appears to follow the excited state temperatures more than the reported ground state temperatures. The remaining codes that model multiple temperatures predict temperature profiles that are more similar in magnitude with the experimental data from the ground state hydrogen measurements and the rotational molecular measurements. The Boyd DSMC model profile is actually quite similar to the data profile observed, while the Subramanium model results appear to predict a lower species temperature in the center of the plume rather than the profile local maximum temperature as is observed in the groimd state experimental data. 4.4.1.3 Nozzle E xit Density vs. m odel results Figure 4-39 shows the calibrated number density profile of hydrogen atoms near the exit plane of the nozzle (0.4 mm from nozzle exit plane). Also shown in the figure are the predicted exit plane number density distributions from the four arcjet models. General agreement between the data and one simulation is observed. The two temperature NS model by Subramanium predicts atomic hydrogen concentrations at similar magnitudes (though slightly lower) as is observed in the experimental data, but with a slightly more peaked profile. The DSMC model also indicates higher number density concentrations than the single temperature NS models predict, though not as high as the Subramanium model, with a maximum of about one-third of what is observed 103 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. experimentally. The DSMC results are not as peaked in profile as the Subramanium data and follow, in trend, the experimental results fairly well. The two other NS models from Butler and Megli both predict significantly lower atomic hydrogen concentrations by almost an order of magnitude. 1 x 1 0 16 i s o > Q I IX IO " 3 z X u s e 1x10 14 # Ground State H data (Pobst) — Subramanium CFD Model - - - - Boyd DSMC Model Butler CFD Model — " Krier CFD Model -6 -2 0 Radial Position (mm) Figure 4-39: Ground state hydrogen density data (previously shown in Figure 4-19) with corresponding data (same arcjet conditions, same profile location) from Navier- Stokes and DSMC Computational m odels. An analysis of the significance of atomic density concentrations follows in section 4.5 and discusses the impact of these results on specific impulse, conservation of mass, and dissociation fraction in the hydrogen arcjet. Additional molecular density measurements from the Raman'' 104 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. work are used in this analysis to determine overall plume densities by combining the results from the two major species in the plume: atomic hydrogen and molecular hydrogen. 4.4.2 ARCJET PLUME AT TWO POWER LEVELS - MEASUREMENTS AND MODEUNG R e su lt s 4.4.2.1 Overview Atomic hydrogen properties were measured in the hydrogen arcjet at two different operating power levels in order to demonstrate that the experimental procedure could be used at multiple thruster operating conditions. Additionally, these physical properties measured at different operating powers offered the opportunity to examine computational model sensitivity to cfianges in the current and voltage of the thruster. The following section compares the data shown earlier with computational results from the Boyd DSMC calctilations at both 1.34 kW and 800 watts.” Additionally, because the DSMC calculations could extend out into the plume, comparisons of experimental data to model predictions out in the plume are provided. 4.4.2.2 Nozzle Exit The following figures show nozzle exit atomic hydrogen properties at both 1.34 kW and at 800 W powers. The figures are grouped such that velocity, temperature, and density profiles may be compared side by side with the higher power case on the left and the lower power case on the right. 105 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.4.2.2.1 Velocitv I > 14 1 2 10 8 6 4 SozxlcEziU 134 kW # H dau ' " DSMC Model 2 0 .2 0 2 Radial Posdon (mm) I 1 I • # Nozzle Exit: 800W # H data DSMC Model Radial Podtkm (mm) Figure 4-40: Hydrogen velocity measurements and DSMC predictions at arcjet nozzle exit for 1J4 kW and 800 W /- Comparing the velocity profiles in Figure 4-40, the 40% decrease in power without change in propellant flow rate appears to have led to a 15% decrease in maximum nozzle exit atomic hydrogen velocity and about a 25% to 30% decrease in velocity at the edges of the velocity distribution. The model prediction for both cases is quite close to the measured velocities predicting somewhat lower velocities in the higher power case and quite closely matching the data observed in the 800 watt plume. For the 800 watt case, the model predicts small increases in velocity near the outer edges of the nozzle walls. While this is not directly observed in the velocity data, the features would be well within the observed velocity scatter and could be present. 106 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.4,2. 2 .2 Temperature 2000 2000 # # # • 1500 1500 ? 1000 g 1000 500 500 Nozzle Ezic: MOW # H data DSMC Model 0 6 2 0 2 Radial Position (mm) 6 4 6 Figure 4-41: Hydrogen temperature measurements and DSMC predictions at arcjet nozzle exit for 134 kW and 800 Comparing the two power cases in Figure 4-41, a overall decrease in average temperature can be seen with decrease in power, though the scatter would make a quantitative measure of the decrease tenuous. For the higher power case, the DSMC model predicts a peak temperature of similar magnitude to that experimentally observed, though the predicted distribution appears to drop in temperatime with radius faster than the observed data does. The scatter in the experimental data is significant enough that it is not able to corroborate the predicted increase in temperature as the radial position approaches the nozzle walls, though it does not disprove the prediction either. The model for the 800W case appears to underpredict the measured temperature but the shape of the profile follows the data as well as can be possibly discerned in the scatter. 107 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.4.2.2.3 Density • • • IzlO 1x10 Nozzle Exit: 800 W • Hdata DSMC Model Nozzle Exit: 1^4 kW DSMC Model # # IzlO 1x10 2 • 2 • 2 0 6 0 2 4 -6 •A 6 -4 4 Radial Position (mm> Radial Position (nun) Figure 442: Hydrogen density measurements and DSMC predictions at arcjet nozzle exit for 1J 4 kW and 800 Reduction of the power by 40% did not appear to significantly impact the number density of atoms at the center of the arcjet nozzle exit. Both atomic hydrogen density peaks appear to be around Ix l0 ‘* cm^ and are maximum at the center of the exit plane. The 800 watt profile in Figure 442, however, appears slightly more peaked than the 1.34 kW profile, falling off an order of magnitude in density within 3.5 mm of center, while the higher power case takes about 4.5 mm to drop to 1x10'^ cm \ Note that the model number density predictions both predict lower values than what was measured, but like the data, the two predictions do not show marked number density changes with changes in power. The profile shape change observed in the experimental data between the two power cases is directly predicted by the model with the more flat profile predicted in the high power case and the more peaked profile shown in the lower power case. As mentioned earlier, the DSMC model^^ uses an upstream initial boundary condition that is derived from the Butler Navier-Stokes 108 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. computational model.* One could surmise that the simulation models the density trend well (as demonstrated in the matching of the profiles at nozzle exit), and an improvement in the upstream start condition might then account for the magnitude discrepancy at nozzle exit. 4.4.2.S 10 mm Downstream The DSMC computational approach is uniquely suited to modeling the plume environment from the arcjet nozzle to the plume region and the following figures explore physical quantities at a downstream radial distribution 10 mm from nozzle exit. These plots show radial distributions about the nozzle centerline 10 mm downstream of the nozzle exit. Again, the figures are grouped by physical property (density, velocity, and temperature) and by power (1.34 kW arcjet operation on the left and 800 W operation on the right). Both experimental data and the DSMC prediction (solid line) are shown in each figure. Additionally, a DSMC prediction is given with the result normalized to the experimental data at the nozzle exit center (dashed line). This provides a comparison of DSMC prediction in the plume region only assuming a start condition for the plume where the model and the data more closely agreed at the center of the arcjet nozzle exit. The prediction of change at a location in the plume relative to the nozzle exit conditions are thus shown in addition to the absolute predictions which take processes inside the arcjet into account. 109 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.4.2.3.1 Velocity I = 10 mm Downstream: 134 kW DSMC Model DSMC: Normalized to Noslc Ejdt data •10 Radial Position (mm) 14 1 2 1 0 8 6 4 10 mm Downstream: 800 W 2 — DSMC Model •••— •DSMC: Normalized to Noole Exit data 0 — •10 • 5 5 0 1 0 Radial Position (mm) Figure 4-43: Hydrogen velocity measurements, predictions, and normalized predictions 10 mm from nozzle exiL^^ The drop in peak atomic hydrogen velocity as power is decreased is almost identical to the drop observed at nozzle exit (from 13 km/s to under 11 km/s), but the distribution is substantially wider at 10 mm downstream. The profile at 1.34 kW appears to drop to about 8 km/s at 10 mm from centerline and 6 km/s at 800 W. This is also a 25% decrease at the wings of the distribution similar to what was seen at the nozzle exit. The velocity profiles from the DSMC model again show slightly lower predicted velocities that almost exactly match up with the data when normalized to the nozzle exit peak velocity. It would appear that for velocity, the change between nozzle exit and 10 mm downstream predicted by the model matches exactly the change observed in the experimental data. 1 1 0 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.4.2.3.2 Temperature 2000 2000 9 mm Downstream: 800 W • H daU DSMC Model - "DSMC: Normilaed to Noizlc Exit 9 mm Downstream: L34 kW • a data DSMC Model DSMC: Nomalizcd to Nozzle Exit 1500 1500 % 1000 g 1000 • X 500 500 1 0 •10 •10 Radial Position <mm) Radial Position (mm) Figure 444: Hydrogen temperature measurements, predictions, and normalized predictions 10 mm from nozzle exit The temperature profiles shown in Figure 4-44 are significantly different at 10 mm downstream than what was observed at the nozzle exit. The profiles of both powers are substantially more flat downstream and while still quite noisy, both exhibit average temperatures of around 700K to 800K at either of the arcjet powers. The DSMC code, however, predicts, strong variations in temperature at the two different powers with a more peaked profile at the higher power and a more flat profile at 800 W. The model prediction normalized to the nozzle exit data does not appear to lead to better agreement between the predictions and the experimental data at this plume location. 4.4.2.3.3 Density Density distributions measured 10 mm downstream of the nozzle exit are shown in Figure 4- 45 and appear to be slightly affected by the change in power. The density maximum in the center of the distribution drops from an average of about 2.4x10** cm * at 1.34 kW to an average of about 1.7x10** cm * at 800 W, though the scatter and sample size make a conclusion on power effect on 111 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. peak density uncertain. At 5 mm from the center, the density does appear to drop as power is decreased from about 6x10'^ cm^ to 4 x i0 ‘ ‘ ‘ cm ", a decrease of over 30%. similar to what was observed at nozzle exit D mm DowBstrtam: 1.34 kW # H data — DSMC Modal —••DSMC: formalized to Nozzk Exit data 1x10' É Î 1x10 I 1x10 5 •S 0 1 0 •10 10 mm DowBstrcam: S O D W # H data — DSMC Model ••••••DSMC: Normalized lo Noafe Exit 1x10 1 a - Ixio' I Z 1x10 s •10 •5 0 10 Radial Positioa (mm) Radial PosiUoa (mm) Figure 4-45: Hydrogen density measurements, predictions, and normalized predictions 10 mm from nozzle exit The model density predictions at 10 mm downstream appear significantly lower than the measured values, but are extremely close in trend and shape. The rate of decrease from the center to the wings matches the measured data almost exactly. The overall magnitude difference already observed at nozzle exit has been compensated for in the dashed prediction and for the high power case the data and normalized prediction match almost exactly. The 800 W prediction appears to predict values lower than the data, but not by as much as the normalization factor, which when applied, causes the normalized DSMC profile to fxedict higher densities than observed. 112 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 .4 .2 .4 A x ia l P ro file Again, the DSMC computational approach allows modeling of the downstream plume environment from the arcjet nozzle out to the plume region. In this section, the following figures explore physical quantities along an axial distribution along the centerline from the nozzle to 30 mm downstream. As in the previous section, dashed lines are added, representing the prediction when normalized to the nozzle exit data. This allows the model to be evaluated in the plume region only. Again, 1.34 kW operation is shown for density, velocity, and temperature in the plots on the left, and 800 W counterparts are shown in the plots on the right. 4.4.2.4.1 Velocity 14 1 2 1 0 I • Î 6 » 1 s • Axial Profile: 1 ItW • H d a u — DSMC Model ......Q S M C : Normalized (o Nozzle Exit data 10 15 20 RadtaJ Position imin) 25 30 1 0 Ï 6 4 Axial Profllc: WOW • H data DSMC Model ......D S M C : Normalized to Sonle Exit data 10 15 20 Axial Position (nun) 25 30 Figure 4-46: Hydrogen velocity measurements, predictions, and normalized predictions along the axial centerline. Figure 4-46 indicates that no significant perturbations to velocity are taking place due to vacuum chamber back pressure interaction with the expansion at these operating pressures. Both velocity profiles show essentially constant velocity, consistent with the centerline values previously 113 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. seen at nozzle exit and at 10 mm downstream. The model predicts that at these pressures the velocities should remain constant, and though lower velocities are predicted, normalization, as would be expected by definition in this case, leads to very close match between model and experiment. The atoms in the plume along the centerline appear to expand as an undisturbed beam of particles. 4.4.2.4.2 Temperature 2S00 2000 E 1000 500 30 Axial Position (mmi 2500 Axial Profile: 800 W • H d a u DSMC Model *'**DSMC: Normalized to Nozzle Exit d a u E 1000 10 15 20 Axial Position (nun) Figure 447: Hydrogen temperature measurements, predictions, and normalized predictions along the axial centerline. Decreases in temperature from nozzle exit into the plume are observed in Figure 4-47. In each plot, temperatures decrease with position from the observed nozzle exit values to somewhat constant downstream values of 700K to 800K. The scatter in the lower power case is again significantly less than that observed in the higher power case. Model predictions for temperature along the centerline exhibit similar decreases. The temperature predictions appear to predict lower temperatures than what the measured values indicate. Normalization of the model results lead the 114 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. trends to match well for the first 10 mm. but then begin to diverge as the data is compared farther out in the plume. 4A.2.4.2 Density IzlO DSMC Model DSMC: Normoiized to Noafe Exit doU T 1 x 1 0 * * 1x10 5 1 0 30 0 1 5 2 0 25 Axial Profile; S O O W # H data — DSMC Model ' — •••DSMC: Normalized to Nozzle Exit data 1x10 1x10 5 0 15 2 0 25 30 10 Axial Podtioo ( mm * Axial Positioo (mm > Figure 4-48: Hydrogen density measurements, predictions, and normalized predictions along the axial centerline. Density for both powers is observed to drop exponentially (somewhat linear on a semi-log plot) though the decrease is slightly faster in for the 1.34 kW case than in the 800W case in Figure 4-48. Noticeably different is the scatter observed at the different powers. The density at the higher power was not as repeatable day to day as was observed at 800 W operation. Though the signal to noise ratio is much better at the lower power arcjet operating condition, no explanation seems completely viable at this point to justify the differences in observed scatter. Note that at the chamber background pressure of 6 Pascals (45 mtorr), no discontinuities in density that would indicate a shock formation are observed in the atomic species density. The Monte Carlo model predictions for the centerline densities indicate similar decreases in density with lower predicted magnitudes. When normalized with the nozzle exit and plotted using 115 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the dashed line, the trend for the 1.34 kW case matches the data well, perhaps predicting slightly higher densities in the far plume. The 800 W case also predicts lower densities than measured, but the normalized slope appears to decrease in density less rapidly than the experimental data indicates. 4.5 A n a ly s is o f N o z z l e E x i t D a t a a t 1.34 k W O p e r a t i o n a n d R e s u l t i n g I m p l i c a t i o n s t o A r c j e t P e r f o r m a n c e P a r a m e t e r s Atomic hydrogen density data is provided in previous sections. As this data is the first to measured at the arcjet nozzle exit, some skepticism as to the accuracy of the absolute number values may be taken since no other atomic density data is available with which to compare. Using computational models for comparison is dicey at best. It has been already shown that atomic density is one thing that the computational models currently under development do not agree upon, and therefore, density is not a predicted parameter that is considered “known.” An attempt may be made to validate the data with the implications drawn from the data with respect to other arcjet performance parameters. For example, specific impulse has been measured for the same arcjet at nearly identical conditions and the implied atomic densities should not contradict this specific impulse measurement. In order to determine the relationship of measured atomic density data to other arcjet parameters, corroboration with molecular density data is required. Only one measurement to date has been made of molecular hydrogen density and this has been done by Beattie and Cappelli at Stanford University.'" Raman spectroscopy was used to measure the molecular density, and due to the high difficulty of the measurement, only five points are available across the nozzle exit for one operating condition. Though more data would be desirable, these points represents the best current 116 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. knowledge of molecular hydrogen density at the arcjet nozzle exit and thus will be used for this calculation. The combination of experimental data can be shown to relate to measured specific impulse, and mass flow data at nozzle exit of the arcjet. Additionally, the molecular dissociation fraction at the arcjet nozzle exit can be determined comparing the number density of the two measurements. 4.5.1 S p e c if ic im p u l s e f r o m d e n s it y a n d Ve l o c it y Me a s u r e m e n t s Total specific impulse can be calculated from integrated measured density and velocity profiles. Additionally, changes in pressure over the expansion can impact specific impulse, but the pressiu'e term is expected to be small and is neglected for tfiis analysis. The total specific impulse can be determined for a hydrogen arcjet, therefore, using the atomic hydrogen density data presented here and the molecular hydrogen density data from Stanford. Both measurements took place at approximately 1.38 kW (1.34 kW for the atomic measurements, 1.4 kW for the molecular measurements), near 13.2 mg/s mass flow (13.1 mg/s for the atomic data, 13.3 mg/s for the molecular data), with similar V-I characteristics for the two arcjets. The integration is done radially from center outward on the positive side of the function. Specific impulse for a rocket engine is given by the following formulation; T Equation 4-1 I„ = ---- ' mg where hp is the specific impulse, T is the thrust, mis the mass flow rate of the propellant, and g is the gravitational constant. The thrust may be rewritten as a function of density and velocity of the propellant gas. 117 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Equation 4-2 T = j p u 'd A A where p is the density of the gas, u is the gas velocity, and dA is the unit of area at the arcjet nozzle exit. Substituting Equation 4-2 into Equation 4-1 results in Equation 4-3 j p u 'd A A where the specific impulse is now given in terms of total mass flow, density and velocity. This can be expanded to include the components of density and velocity and the drop in pressure between the thrust chamber and the ambient atmosphere which is shown in the following relation. Equation 4-4 I = mg \p„Ufi'-dA + \ { p - P o )àA \ A where pn is the density of atomic hydrogen, pœ is the density of molecular hydrogen, uh is the velocity of atomic hydrogen, uhj is the velocity of molecular hydrogen, p is the thrust chamber pressiue and po is the ambient pressure. As the experimental data available is that of number density rather than mass density, the equation can be rewritten to accommodate the use of the measurements. 1 Equation 4-5 I = ----- mg jnH m „u„ -dA + + j ( p - Pa)dA \ A A A where nn is the atomic hydrogen number density, nm is the molecular number density, nin is the mass of a hydrogen atom, and mn2 is the mass of a hydrogen molecule. The third pressure difference integral term in Equation 4-5 is considered negligible with respect to the velocity weighted momentum of the species and can be dropped. This allows the U p to be written in terms of the species densities and velocities which have been measured. 118 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1 ^ Equation 4-6 I = mg \A This can be rewritten for purposes of symmetrical integration into the following relation 2 ^ ^r=4..5mm r=4jm m '' Equation 4-7 ^:p ~ + j>^h2'^ h 2Uh2'rd r V '•=0 r=ü J where dA is calculated radially z s r dr dd, with r going from 0 to 4.5 mm (the radius of the nozzle exit) and 6 going from 0 to 27t. For the calculation, nnix) is determined from a curve fit of the measured ground state atomic density data at 1.34 kW (see Figure 4-49). Three curve fits were made on the atomic hydrogen density profiles representing high, low, and average values of the data scatter. This was done to determine the deviation in mass flow and Isp due to uncertainties in the density profile. The molecular density profile ««(x) for purposes of the integration is determined using a polynomial curve fit of the Raman molecular density data taken by Beattie and Cappelli'* at 1.4 kW (see Figure 4-50). The molecular hydrogen densities are known ± 25% and high, low and average values were used to determine the deviation in mass flow and Isp due to uncertainties in the density profile. 119 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ? i I I E 3 z 16 1x10 15 1x10 u 1x10 ■ 6 -4 -2 0 2 4 6 Radial Position (mm) Figure 4-49; Atomic hydrogen density data as seen in Figure 4-19 with polynomial fits for high, low, and mean density values. ? S w 1 u s 3 z 16 1x10 15 1x10 14 1x10 -6 -4 2 0 2 4 6 Radial Position (mm) Figure 4-50: Molecular hydrogen density data reprinted from Stanford University with the addition of a polynomial fit and error bars, the latter are placed as described in the technical report of the measurement‘s 1 2 0 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Velocity of the atomic species has been measured in a variety of ways with good agreement across the different techniques. For purposes of this calculation, uh(x) is determined from a curve fit of the atomic ground state data at 1.34 kW. This velocity profile has been shown in Figure 4-37 to be essentially identical to excited state velocities measured by Liebesldnd, Hanson, and Cappelli [1992]''* (see Figure 4-51). The velocity of the molecular species uh2(x) is assumed to be the same as Uf](xj. This is expected to be a sound assumption based upon measurements determining little or no difference in velocity between excited state atomic hydrogen and helium (which has a similar mass to the molecular hydrogen) seeded into the flow in Liebeskind, Hanson, and Cappelli [19931. 38 3 ë 4 ) > 14 12 10 8 6 4 2 1 5 6 0 0 2 -6 -4 -2 4 Radial Position (mm) Figure 4-51: Atomic hydrogen velocity data as seen in Figure 4-12 and polynomial fît of mean velocity. 121 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. As the flow rate was 13.1 mg/s for the atomic density experiment and 13.3 mg/s for the molecular density experiment an m of 13.2 mg/s will be used for the calculation. The results of the integration provide an Isp component due to each species which when added together comprise the total specific impulse. The uncertainties in density also lead to corresponding uncertainties in specific impulse. The implied specific impulse and deviation for each of the experimental data sets is summarized in Table 4-1 below. Atomic hydrogen measurements (2PLIF) Molecular hydrogen measurements (Raman) Combined Power Spec. Pwr. 1340 W 102.3 1400 W 105.3 low 178 Implied Isp of H (s) average 333 high 487 low 455 Implied Isp of H2 (s) average 606 high 758 low 633 Implied Total Isp (s) average 939 high 1245 Table 4-1: Implied Isp values for each density measurement for a low power hydrogen arcjet Using the mean values of each of the data sets, the implied Isp from the two density measurements is 939 seconds. The specific impulse for these exact operating conditions has not been measured using thrust stand measurements, but specific impulses in similar arcjet operating conditions were measured by Pencil et al at the NASA Lewis Research Center.” Operation and specific impulse measurements (including implied Isp’s from each experiment for it’s species component of the total Isp) are tabulated below in Table 4-2. 1 2 2 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Atomic hydrogen measurements (2PLIF) Molecular hydrogen measurements (Raman) NASA Lewis (Pencil et al) NASA Lewis (Pencil et al) Electrode Gap (mm) 1.78 7 1.52 1.52 Mass Flow (mg/s) 13.1 13.3 15 10 Voltage (V) 134 140 128 113 Current (A) 10 10 11.7 11.1 Power (W) 1340 1400 1498 1254 Spec. Pwr.(MJ/kg) 102.3 105.3 99.9 125.4 Isp due to H (s) 333 - - - Isp due to H2 (s) - 606 - - total Isp (s) - - 861 883 Table 4-2: Arcjet Operation and Specific Impulse Measurements The experimental diagnostic conditions for the atomic and molecular hydrogen measurements fall somewhere between the two Lewis thrust stand measurement operating conditions shown in Table 4-2. and therefore, it is expected that had the mean operating condition of the two density experiments been tested at Lewis, a specific impulse in between the two Lewis cases, such as 870 seconds would likely be measured. This estimate is based upon examination of the two NASA Lewis operating conditions. The 1498 watt case had a higher mass flow than that of the optical experiments causing the Isp to be lower than what would be measured at the lower mass flow case. The 1254 watt case would likely have a greater specific impulse than the hydrogen measurements average condition as the specific energy was higher and the flow rate lower. A specific impulse value in between the two cases is an acceptable guess in light of the absence of a thrust measurement at the operating conditions where the two hydrogen measurements were taken. So, the combined diagnostic Isp of 939 seconds is found to be about 8% higher than that measured at NASA Lewis. Examining the range of Isp's implied by the measurements, it is seen that the experimental deviation in Isp (shown in Table 4-1) can be as high as 33% and, therefore, 123 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the 8% difference is well within the experimental uncertainty. Note that if the atomic hydrogen density was assumed to be at the lower range of the experimental scatter (i.e. peak center density is 7 X 10‘^ cm^ rather than 1 x 10‘* cm'^) the total implied Isp (assuming the average molecular density values ) would be 784 seconds or 11% lower than measured Lewis specific impulse numbers. This would imply that the range of atomic hydrogen densities measured by Pobst (in conjunction with the molecular Raman data taken at Stanford) is quite consistent with measured specific impulse data. If actual atomic densities were lower than those measured, the deviation from measured thrust stand data would increase. The differences between the specific impulses implied by the density experiments and the specific impulses measured via thrust stand are small enough that the deviation cannot be attributed directly to either density measurement. 4.5.2 Co n se r v a tion o f To tal Ma ss f l o w A similar exercise to the one in the previous section for total mass flow and how the density measurements account for total mass flow is warranted as well. Integration of the density and velocity distributions are conducted in a similar fashion as above with mass flux calculated rather than specific impulse. Equation 4-8 rh = j p„UffdA + j dA A A Equation 4-9 m = j 4- j / r=4.5mm r=4.5mm Equation 4-10 m = 2n \n„ m „ u„ rdr+ \n fj,m ^.u „ .jd r V r= 0 r=0 > The results of this integration (separated by species term to coincide with each experiment) are shown in Table 4-3. 124 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Atomic hydrogen measurements (2PLIF) Molecular hydrogen measurements (Raman) Combined Power Spec. Pwr. 1340 W 102.3 1400 W 105.3 low 2.1 Implied m of H (mg/s) average 3.8 high 5.7 low 6.2 Implied th of H2 (mg/s) average 8.3 high 10.4 low 8.3 Implied Total th (mg/s) average 12.1 high 16.1 Table 4-3: Implied mass flux values for each density measurement for a low power hydrogen arcjet Again, the sum of the average implied mass fluxes 12.1 mg/s is near the average experimental value of 13.2 mg/s and is 9% lower than expected. The experimental error due to density measurements (the error in velocity is trivial for this exercise) is about 33% as before so the agreement with expected mass flow is quite good. 4.5.3 M o lec u la r D is s o c ia t io n F r a c t io n a t the a r c j e t N o z z l e E x it Integration of total atomic and molecular number densities over the arcjet nozzle exit from these measurements can also be done (the number density derivations are quite similar to the two previous sections concerning thrust and momentum, and will therefore be neglected in this section). Performing the integration, the total number of hydrogen atoms leaving the arcjet nozzle exit is about 3.4 x 10‘® m ‘ for the average atomic density profile. The atomic number density represents half that number of original molecules or 1.7 x 10‘® m ‘. The total number density of molecular hydrogen (integration of the Raman density measurements) is about 4.5 x 10'® m ‘. Table 125 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4-4 examines the sensitivity of this percentage showing minimum and maximum values based upon the experimental imcertainty for each measurement. Atomic hydrogen measurements (2PLEF) Molecular hydrogen measurements (Raman) Combined Power Spec. Pwr. 1340 W 102.3 1400 W 105.3 low 1.8 X I0‘® Implied Number/m of H average 3.4 X 10'® high 5.0 X 10'® low 3.4 X 10'® Implied Number/m of H2 average 4.5 X 10'® high 5.6 X 10'® Implied Pre-disassociated low 4.3 X 10'® Total Number/m of H2 average 6.2 X 10'® high 8.1 X 10'® Table 4-4: Implied total number values for each density measurement for a low power hydrogen arcjet The effect of the uncertainty on the total dissociation fraction is shown in Table 4-5. The dissociation can be predicted from these two measurements to be as low as 14% using the lower boundary of measured atomic densities and upper uncertainty of molecular densities and as high as 42% when using the upper boundary of atomic measurements and the lower uncertainty of the molecular measurements. The overall percentage of molecules subject to dissociation using the average density values mentioned above is about 27 % for the nozzle exit profile. 4.5.4 ENERGY L o s s DVE TO MOLECULAR DISSOCIA TION IN THE HYDROGEN ARCJET The dissociation of molecules to atoms takes energy away from the production of directed thrust. Once a dissociation fraction has been estimated, a determination of how much energy is used for the dissociation process can be made. Using the average measured value determination of 126 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 27% dissociation for the energy calculation, one can calculate the corresponding energy required to break apart the hydrogen molecules that provide the measured number of atoms in the flow. Atomic Prediction (1 X 10'" /m) Implied Pre disassociated molecules (1 X 10‘®/m) Molecular Prediction (Ix lO'Vm) Total Molecular Number Density (Ix lO'Vm) Disassociation Fraction 1.8 0.9 5.6 6.5 14 % 1.8 0.9 4.5 5.4 17 % 1.8 0.9 3.4 4.3 21 % 3.4 1.7 5.6 7.3 23 % 5.0 2.5 5.6 8.1 31 % 3.4 1.7 3.4 5.1 33 % 5.0 2.5 4.5 7.0 36 % 5.0 2.5 3.4 5.9 42% Table 4-5: Range of predicted disassociation fractions using the high, low, and average values of measured atomic and molecular hydrogen densities at the low-power arcjet nozzle exit. The predicted disassociation for the average measured values is highlighted in gray. The number flux of dissociated molecules (as opposed to the number of atoms, where there are two per dissociated molecule) passing through the arcjet can be represented by the following expression; Equation 4-11 = 'molecules M where is the number of molecules that were broken apart, ^ is the measured dissociation fraction, is the hydrogen gas flow rate, and A / « 2 is the mass of the hydrogen molecule. Using the 27% value for dissociation fraction derived earlier, and the average flow rate of 13.2 mg/s of hydrogen, the number of dissociated molecules per second, 1.07 x IQ-' molecules per second. 127 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Dissociation energy of the hydrogen molecule is given to be Ed^ = 4.478 eV per molecule.'^ Assuming each dissociated molecule per second requires this energy, the power required for sustained molecular dissociation can be determined: eV Equation 4-12 = 4.8 x 10‘‘ — = 770 Watts The power requirement of dissociation as determined from the two density experiments is 770 Watts out of an average power of 1370 Watts or 56% of the energy. While this number appears at first to be larger than might be expected it is also consistent in magnitude with predictions of other energy expenses. Aerospace Corporation estimates for energy losses in their arcjet thruster can be used to help determine a total energy prediction. While Aerospace Corp. uses an identical model arcjet to the one in this study, it has been operated at a much lower flow rate and somewhat different operating conditions. The estimates for their thruster operation, however, should not be significantly different from the thruster operation in this study and can aid in this order of magnitude energy balance exercise. Measured radiant heat loss in a hydrogen arcjet from these tests range from 9% to 11% of input energy at powers ranging from 1 kW to 1.2 kW, and random heat addition to the propellant is expected to be of order 0.1 eV per molecule.'*^ VUV measurements predict 0.12 eV per molecule for vibrational energy losses and 0.035 eV per molecule for rotational energy loss."* For the 13.2 mg/s flow at 1370 Watts, this would amount to about 10% of the energy in radiant heat, 3% of the energy in random heat addition to the propellant, 4% of the energy in vibration, and 1.2% energy trapped in rotation. Ionization is predicted to be n e g lig ib le ,s o energy loss due to ions and electrons are expected to be insignificant for the purposes of this estimate. Using the derived 56% 128 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. amount of energy for dissociation and a educated estimate (based on the performance of other similar arcjets)” o f near 30% efficiency for this laboratory model arcjet, the summation of these losses: E()UatlOn 4* 13 E Ethm st (Eradiate"^ E hcat addiiion~^ Evibracion'^ EroiaQon"^ E dissocubon) = 30% +(I0%+ 3% + 4% + 1.2%+ 56% ) = 30% + (74.2%) = 104.2%. \\7iile totaling to slightly greater than 100% of the energy provided to the arcjet. the energy summation appears to be a reasonable one. The order of magnitude of a 56% dissociation energy, as determined through the direct measurement of hydrogen atoms and hydrogen molecules, is not inconsistent with the other predicted energy losses, and whether exactly 56% or not, it would appear that the energy loss to dissociation of hydrogen molecules is likely the most significant energy loss mechanism in the arcjet thruster. 129 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 5 In Conclusion This work demonstrates the adaptation of the ground state hydrogen diagnostic tool demonstrated by experimentalists in the flame community to a new application in the plasma plume of an electric propulsion thruster. Application of this laser approach to the arcjet environment involves additional exercises and modifications in order to perform the technique in a manner that yields ground state hydrogen densities. Because these modifications were successfully implemented, the first measurements of ground state hydrogen atomic densities were not only successfully performed, but spatially mapped in a detailed manner at the arcjet nozzle exit and downstream in the plume. This approach also yielded temperature and velocity data that in the case of velocity supported previous experimental measurements, but in the case of temperature showed the potential pitfall of using the excited state temperatures as representative of the entire flow, as the corrections for Stark broadening in the excited states can be complex and significant. The density results caused a significant commotion in the community as these first atomic density measurements recorded did not corroborate the predicted and expected concentrations, and in fact, demonstrated significantly higher amounts of atomic hydrogen than expected, indicating that the frozen flow losses due to dissociation are also higher than had been expected. These higher concentrations are shown to still be consistent with other molecular density measurements that have been conducted as well. In fact, no indication is present that either experiment is overpredicting or underpredicting species densities. Rather, the agreement of specific 130 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. impulse and mass flux to that implied by the sum of the density measurements further strengthens both data sets as accurate depictions (to within their respective experimental uncertainties) of arcjet flow properties. These experimental results, have, therefore, significant implications on how much improvement in performance and efficiency can be accomplished with additional redesign of the arcjet, and on what type of improvement approach will be required if gains are to be realized. Additionally, this work was able to be compared to several different modeling approaches, allowing insight into the different approaches with respect to how they predict atomic hydrogen concentrations. Experimental data was taken at multiple arcjet operating power and at multiple locations in order to provide several possible comparisons with different arcjet modeling approaches. A large database of atomic hydrogen density, temperature, and velocity has been created using the developed technique. Community acceptance of this approach has also be favorable. Several different researchers are now contemplating extending this technique to Xenon in order to investigate Hall-effect thruster plumes, and a paper including some of the work presented in this thesis has been chosen as “Best Electric Propulsion Paper" of the 1996 AlAA Joint Propulsion Conference by the Electric Propulsion Technical Committee. In conclusion it is believed that this effort has successfully accomplished each of the goals previously set out in order to investigate the ground state hydrogen atoms in the arcjet plume. 131 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1 Literature Review Appendix 1.1 Re s e a r c h o n A r c je t T h r u st e r s Arcjet thruster research has been ongoing at various levels of effort since the late I950's. Recent advances in spacecraft power available and power processing unit circuitry have enabled a renewed interest in the use of electric propulsion for space missions. Use of these devices in many missions is considered desirable due to the large increases in available specific impulse found in electric thrusters when compared to chemical thrusters. For many missions, arcjets are considered the most developed electric propulsion device, and implementation of them on board spacecraft is already beginning. Though certain arcjet models, (such as the Olin Aerospace Corp. 1.8 kW hydrazine arcjet) are considered flight-qualified and perhaps flight-proven, basic physical processes in the arcjet operation are far from understood. The extreme inefficiency of these devices (only as high as 35% efficient) suggests tfiat significant room for improvement is available through redesign based upon understanding of the fundamental processes that convert the applied electric power to directed thrust. Even small gains in efficiency lead to a significant increase in the number of missions where arcjet use becomes advantageous. Because of this, many groups have been and are now involved in exploring the physical behavior of the arcjet thruster. The following sections describe the work in tfiis area being conducted at research facilities around the world and are listed in order of contribution as measured by number of publications on arcjet research during the last decade. 132 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. l.I .l NASA LEWIS research CENTER The Lewis Research Center of the National Aeronautics and Space Administration has one of the world’s largest satellite propulsion programs. It includes an active electric propulsion program that has specialized on ion engines and arcjet thrusters among others. During the last ten years, NASA Lewis has developed an aggressive arcjet program and has helped standardize research conditions on low-power or 1 kilowatt (kW) arcjets by providing several laboratory model thrusters and power supplies to universities and research laboratories throughout the United States. In addition to equipment and financial support for outside research, Lewis’ internal research has addressed many technical and system level issues on the arcjet thruster. Researchers at Lewis have investigated arcjet design parameters by testing performance while adjusting electrode gap spacing, constrictor diameter sizing, nozzle area ratios, nozzle geometry designs, forced arc attachment locations, and propellant flow ranges for hydrogen, nitrogen, simulated ammonia, and simulated hydrazine arcjets at the 1 kW, 10 kW, 30 kW, and sub kW power classes. The Lewis group has designed and built power electronics for arcjets at each of these ranges as well In order to test performance, the Lewis electric propulsion group has designed arcjet thrust stands for the various power levels and examined their use in different chamber pressure environments and at different thruster operating conditions. NASA has also performed experiments on arcjet performance loss mechanisms through examination of cathode erosion, supersonic and subsonic arc attachment phenomenon, arc restrike interference, and generalized thermal radiative losses. The electric propulsion group has also made efforts to enhance performance with designs for regeneratively cooled nozzles, methods for low erosion starting, new 133 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. materials for electrode construction, and an attempt to implement a high pressure hollow cathode. Some small increases in performance have been realized with these methods. Lewis’ EP group has provided support for spacecraft flight programs through mission analysis, arcjet system life tests, electromagnetic interference tests, flight power electronics design, arcjet ignition reliability, plume contamination studies and cathode life tests. The Lewis EP group has conducted some experiments to determine basic physical arcjet plume parameters including the use o f electrostatic probes to measure electron density and electron temperature in hydrazine and hydrogen arcjets. Emission spectroscopy on N2, N2+, NH, and H have also been performed on ammonia arcjets with the Stark linewidths and the Balmer series of H being used to determine electron densities and temperatures inside the arcjet nozzle (through two access holes) and in the exterior plume regiort Significant non-equilibrium distributions were observed. In all, over fifty papers, journal articles, and reports have been published by the NASA Lewis electric propulsion group representing almost one fifth of the arcjet literature published in the last twenty years. 1.1.2 ROCKET R e s e a r c h C o m p a n y / O u n a e r o s p a c e C o m p a n y / p r im e x A e r o s p a c e C o m p a n y 33.93.1201 Olin Aerospace Company (formally Rocket Research Company) is the only manufacturer of flight qualified arcjet hardware in the United States and a major supplier of other spacecraft propulsion systems (Olin has recently changed their name again to Primex Aerospace Company, but is consistantly referred to in this document as Olin). Olin has participated in many research 134 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. programs involving the electric propulsion community and has published results from many internal arcjet research programs, computational and experimental. Olin has conducted complete flight qualification programs for two arcjet thruster systems. A 1.8 kW hydrazine arcjet system has been qualified for spacecraft stationkeeping purposes, and a 30 kW ammonia arcjet system has been qualified for flight on an experimental Air Force satellite in order to demonstrate orbit transfer capability with electric propulsion. Flight qualifications consisted of life testing, propellant system development and qualification, performance mapping, power control unit development and qualification, interface qualifications, electromagnetic interference measurements, thermal and vibrational qualifications. Related work on hydrogen arcjet systems has also taken place. Performance enhancing experiments at Olin, have consisted of projects involving the implementation of regeneratively cooled anodes, subsonic arc attachment designs, and constrictor/electrode adjustment, all to improve heat transfer to propellant. Ignition stability projects have also been undertaken to insure consistent arcjet startup. Olin has also contributed to research on better physical understanding of arcjet thruster behavior. Experiments measuring electron density and electron temperature have been conducted through the use of emission spectroscopy and excited state species temperatures of the first nine lines of the hydrogen Balmer series have also been measured. Arc stability and ignition have been examined in order to better understand the behavior of the electric arc inside the thruster. A coUisional-radiative model was developed to help understand the non-equilibrium observed in the temperature measurements and a full two-dimensional axisymmetrical computational model describing the fluid behavior in the arcjet has been written. 135 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Olin’s arcjet computational model is one of the most mature in the field and it is expected that this model is currently used for preliminary arcjet design. Initially developed as a simple single fluid mass and energy transport model for low power hydrogen arcjets, the Navier-Stokes based model has been improved to support two-fluids, multiple power and geometries, species diffusion, ohmic heating, and an advanced thermal anode mapping solver. One of the strong points of this modeling effort is the interaction between Olin Aerospace Corp. and Stanford University {see section 1.1.8 below). Stanford velocity and temperature measurements of the excited state hydrogen population at thruster nozzle exit have been used to validate and improve the predictive capabilities of the model. Validation of the predicted density profile has yet to take place as no species density measurements (other than electron densities) are currently available. 1.1.3 AIR F o r c e P h il u p s La b o r a t o r y / U n iv e r s it y of S o u t h e r n Ca u f o r n ia (R e f e r e n c e s: 13,35,121-1401 The Air Force Phillips Laboratory (formally the AF Rocket Propulsion Laboratory, and also the AF Astronautics Lab) has been involved in electric propulsion research for several decades and is noted as the place where resistojet and pulsed plasma thruster (PPT) technology were first developed. In the last ten to fifteen years the laboratory has specialized in resistojet thrusters, high power magnetoplasmadynamic (MFD) thrusters and high- and low-power arcjet thrusters. The high-power arcjet program at Phillips Laboratory (PL) was developed to demonstrate orbit transfer with electric propulsion. The Electric propulsion Space Experiment (ESEX) is a flight demonstration of a 30 kW ammonia arcjet operated in low earth orbit on board the Advanced Research and Global Observation Satellite (ARGOS). The flight program has led to many 30 kW 136 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ammonia arcjet experiments on thruster life, thermal radiation and conduction, cathode erosion, electromagnetic interference (EMI), thruster performance ranges, plume contamination, and ammonia propellant handling. Fundamental physical research in propulsion has also been conducted at the Phillips Laboratory on high power arcjets including electron number density and temperature measurements in the plume and inside the thruster nozzle using emission spectroscopy of the hydrogen Balmer series and the NH vibrational lines. Ignition research has also been conducted on the initiation of electric breakdown in the arcjet nozzle and the relationship of the rate of voltage increase during breakdown to sustained and reliable arcjet operation. Magnetic field lines inside and outside of high power arcjet flight power supplies have also been explored by PL using probe techniques. Laser Induced Fluorescence (LIF) of the excited state hydrogen in a high power arcjet was done jointly by researchers from the University of Southern California (USC) and researchers at the Jet Propulsion Laboratory (see section 1.1.4) . This initial experiment determining excited state temperatures and velocities led to a multi-year collaborative effort between USC and the Phillips Laboratory to explore arcjet diagnostics for measuring fundamental physical parameters of the arcjet plume (papers from USC are included in this section as all but that first one are co- authored with PL personnel and all focus on arcjet diagnostic techniques). This low-power ( 1 kW') hydrogen arcjet program consisted of experiments featuring emission spectroscopy measurements and an investigation of arcjet - power electronics interaction with regards to propellant heating, velocity changes, and efficiency. A bulk flow velocity diagnostic was developed utilizing a microsecond scale current pulse to the applied arcjet current with observation of a increased excited-state species population propagating through the observable 137 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. plume. This provided lime of flight information for determining velocity. Facilities at the Phillips Laboratory allowed this velocity diagnostic to also be used to characterize vacuum chamber background pressure with arcjet plume measurements so as to provide insight into acceptable operating pressure ranges for ground test plume measurements. Temporal changes in arcjet electron density and temperature were measured using triple langmuir probes and demonstrated the sensitivity of the arcjet plume to the ripple of the applied current from the arcjet flight power supply. Temporal changes in velocity with respect to the power processing unit were also explored, but found to fluctuate without direct reference to the input current power supply ripple. Efforts to initiate a pulsed electron beam fluorescence diagnostic in an arcjet plume to measure species densities were also begun as density measurements are needed to compliment the velocity measurements in order to identify mass flux plume properties and compare with computational models in the electric propulsion community. PL has also conducted several mission analysis studies on payload orbit transfer (low earth orbit to geostationary orbit), stationkeeping, upper stage replacement and satellite repositioning. 1.1.4 N A SA J e t P r o p u l s io n L a b o r a t o r y The Jet Propulsion Laboratory (JPL) has been conducting research in electric propulsion for many years. Recent research has focused on high power arcjets, ion engines, magnetoplasmadynamic (MPD) thrusters, stationary plasma thrusters (SPT), and anode layer tfmisters (known by their Russian inventors as thrusters with anode layer or TAL ). Arcjet research at JPL has focused on high power ammonia arcjets and largely has been in support of the ESEX flight experiment (described in section 1.1.3) or another Air Force satellite program (ELITE) that was canceled prior to construction. 138 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Experiments at JPL consisted of tests on 30 kW and 10 kW arcjet thrusters in the areas of performance, length of life, system packaging, materials analysis, cathode tip geometry design, cathode whisker growth phenomenon, cathode life testing, power processing unit design, propellant options and their ramifications, contoured nozzle designs, and varied nozzle geometries. Physical measurements on arcjet plumes include axial velocity measurements through the measurement of Doppler shifts via emission spectroscopy and additional use of Fabry-Perot spectrometers. Analysis of emission lines and their strengths has taken place. A specialized facility at JPL also allowed detailed thermal mapping of a cathode geometry similar to that of the arcjet cathode. Computational analysis and modeling has been conducted at JPL with a detailed analytical cathode model, orbit transfer mission analysis and orbit angle of insertion calculations comprising much of the computational work done during the past decade. 1.1.5 INSTITUT FÜER RAUMFAHRTSYSTEME AT UNIVERSITAT STUTTGART (GERMANY) [R e f e r e n c e s : 162-I79[ The electric propulsion program at the University of Stuttgart is a focused program that includes experiments that determine fundamental physical parameters of arcjet plumes and programs that support space flight experiments with arcjet propulsion systems. Researchers at IFR have examined arcjets of all power levels (1-100 kW). Many different arcjet designs have come out of Stuttgart both water cooled and radiation cooled at a variety of power level classes and sizes. Most experiments have used simulated hydrazine as the arcjet propellant, though a few have used hydrogen alone, nitrogen alone, and simulated ammonia 139 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Many plume measurements have been made on 1.5 kW class arcjets on simulated hydrazine. Electron density and temperature have been investigated using emission spectroscopy and attempts at determining local point measurements through Abel inversion and modifications of Abel inversion techniques have been undertaken. NH and H have both been observed with the excitation temperatures of NH vibrational and rotational lines also being measured both in the arcjet plume and at two locations in the interior of the arcjet nozzle. Estimates of species density have been inferred based upon measured populations of excited state hydrogen and NH leading to generalized estimates of frozen flow loss in the arcjet. A two dimensional CCD camera was used to attempt to ascertain electron densities via Stark widths without the use of Abel inversion. A window was placed on the arcjet thruster into the constrictor region and images were taken through the window. The two dimensional profiles of the first two Balmer series intensity lines were used to characterize the gas flow and compare it with the temperature profiles obtained from the electron continuum in the arc region itself. These measurements indicated thermal equilibrium in the arc region and thermal nonequilibrium outside of the arc region leading to the development of an analytical model describing the electron temperature transition between the two regions. CCD measurements were also taken at a variety of operating conditions and compared with pyrometer and thermocouple nozzle measurements to help develop a finite element temperature model that describes the electronic, vibrational, rotational, and anode body temperatures as a function of operating conditions. The model demonstrated a tiigh degree of non-equilibrium. An arcjet performance program is also taking place at IFR and tias thus far consisted of many thrust stand performance measurements on a variety of arcjet designs. Changes in cathode gap spacing, constrictor diameters, electrode materials, and propellant injection methods have all been 140 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. explored Measurements of current distribution, cathode erosion rates, chamber background pressure, and propellant feed pressure have all been conducted and related to overall thruster performance. Flight qualified power electronics and propellant systems have also been developed at IFR. 1.1.6 A e r o s p a c e C o r p o r a t i o n • '> The Aerospace Corporation created a laboratory dedicated to the examination of electric propulsion devices in 1985. In the last ten years, this lab has developed diagnostic techniques for use in arcjet and ion engine plumes. Having several researches emphasize diagnostic techniques in arcjet plasma environments has led to greater ability to measure previously unknown plume parameters. One of the most involved diagnostics that has been developed at the Aerospace Corporation is their custom designed mass spectroscopy velocity analyzer for examining the species in the far field arcjet plume region. Atomic and molecular species from a 1 kW hydrogen/nitrogen mix arcjet plume were measured in quantity and energy at various angles to the direction of the impinging plume. Arcjet plume species using other propellants were also examined and species density with respect to angle differed greatly between propellant choices. Not only are these measurements of value in determining quantity of species surviving in the far field, but concerns of contamination onto the spacecraft from the propulsion device are directly examined with this technique. Optical diagnostic experiments on arcjet plumes have included emission spectroscopy of rotational and vibrational temperatures of NH in an ammonia arcjet, emission spectroscopy of atomic hydrogen (and separately helium) to examine excited state temperature along with electron temperature and density. Additionally, VUV absorption of ground state hydrogen atoms was 141 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. performed to attempt to ascertain number density out in the far-field plume. Absorption measurements were also conducted on ground state NH to determine number densities and ways excited state NH is repopulated. Laser induced fluorescence (LIF) was used to measure atomic hydrogen excited state temperatures, and axial, radial and azimuthal velocities. Coherent Anti- Stokes Raman Spectroscopy (CARS) velocity and temperature measurements have also been conducted in resistojet plumes with proof of concept measurements demonstrated in arcjet plumes. Support of the Phillips Laboratory’s ESEX program {see section 1.1.3) has led to electromagnetic interference (EMI) testing on 26 kW ammonia arcjets, orbit transfer studies, and mission analysis trade studies at Aerospace Corporation. Additionally, electric propulsion technology readiness studies and studies examining the applicability of electric propulsion to various missions have been conducted. A study indicating how a diagnostic program could best benefit the development of an advanced arcjet thruster was also carried out. A unique program examining the viability of using helium for arcjet propulsion not only from a performance perspective, but from a system wide point of view is also taking place at the Aerospace Corporation. 1.1.7 BP D DIFESA E SPAZIO (ITAL Y) Flight development of arcjet thrusters for flight on European satellites has been a main focus of the electric propulsion program at BPD. Many thruster designs have originated at BPD with testing programs complimenting each design. Arcjet thrusters at 1 kW and 10 kW power classes have been designed, developed, and tested on hydrazine, ammonia, and hydrogetL Power conditioning units have been developed for each thruster class with life tests, performance mapping, and system integration tests conducted for each 142 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. thruster design as well. Often, multiple models of a new design were tested and tests conducted on each model for redundancy. Mission analysis and propulsion system studies have also been conducted to further mature the technology. An emission spectroscopy experiment was conducted at BPD to examine excited state temperature and infer electron number density. 1.1.8 S ta n f o r d Un iv e r s it y The Mechanical and Aerospace department at Stanford University has been active in arcjet thruster research exploring optical diagnostic techniques that provide information on arcjet plume physical parameters. In addition, a substantial diamond film program using arcjets for diamond production is ongoing at Stanford. Many emission measurements have been made at Stanford including examination of the hydrogen arcjet plume in order to determine electron number densities and excited state temperatures both in the plume and in the arc constrictor region in front of the cathode. Abel inversion techniques assuming concentric “shells" have been used to infer spatial excited state temperature and electron density axial and radial profiles. Solid body emission measurements have explored the gray body emission of the cathode and anode to determine the temperature distributions of the electrodes and how close to the melting point of the material (usually 29c thoriated tungsten) that the temperatures approach. Stanford followed up on the laser induced fluorescence (LIF) work of Erwin, et. al. (see section 1.1.3) by using continuous wave (CW) argon-ion pumped dye lasers on 1 kW hydrogen arcjets (rather than a NdiYAG pumped pulsed dye laser on 30 kW ammonia arcjets) to also promote the first excited state of atomic hydrogen to the second excited state (n=2 to n=3) and view 143 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the Doppler widths and shifts of the resulting fluorescence. Because of the completeness and accuracy of the resulting excited state velocity and temperature information, the experimental data became an important factor in the development of models from Olin Aerospace Corp. (see section 1.1.2) and others. Later, the temperature data from the LD F was found to be interpreted as hotter than it should due to a new understanding of the Stark broadening component o f the measured linewidth of the 2 to 3 transition. Concern over the meaning of excited state velocity measurements representing the velocities of the more prevalent atomic ground state species led to an experiment probing a flow of hydrogen with a trace of helium and examining fluorescence from both species. Velocities of the excited states of each were compared and found to be nearly identical. While this does not directly address the ground state vs. excited state hydrogen species concern it does indicate that the probability of ground state hydrogen being substantially different in velocity from excited state hydrogen is likely to be quite small. An attempt was made at determining the density of the ground state atomic hydrogen using absorption spectroscopy in the ultraviolet region. Unfortunately, optical depth problems led to the conclusion that single photon absorption or excitation would be ineffective without the prior knowledge of the spectral line shape of the absorbers. Consequently, it was determined that atomic hydrogen density measurements were not possible using this technique. In order to better understand the molecular component of the hydrogen arcjet plume, a diagnostic program examining Raman scattering was begun. Initially experiments were performed on cold flow through the arcjet nozzle in order to test out the viability of the diagnostic. Recently, attempts at measuring molecular hydrogen densities, temperatures, and pressures in a lit arcjet have also been attempted. Comparison of the cold flow measurements with Direct Simulation 144 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Monte Carlo (DSMC) measurements conducted at Cornell University (see section L I. 15) led to insights on the contribution of the experimental chamber back-pressure to the flowfields that are examined by arcjet diagnostics. Further work to improve the code to model lit arcjets is discussed in a later section. Stanford has also created some analytical models to help explain the experimental results from the programs detailed above. An electron density model and a one-dimensional radiative transfer model were designed to help interpret emission results in the plume. An arc attachment model from cathode to anode in the presence of a flowing thermal plasma was also written in order to help explain the arc attachment modes and explain data taken when examining the arc/constrictor regions. 1.1.9 UNIVERSITY OF TENNESSEE SPACE INSTITUTE 223-232] UTSI has approached arcjet research both computationally and experimentally. Development of computational arcjet simulations preceded the initial diagnostic experiments which were mainly comprised of optical emission and laser fluorescence techniques. Computational simulations on 1 kW ammonia arcjets have been developed and compared (with some disparity) to performance results from the Institut Filer Raumfahrtsysteme at the University of Stuttgart in Germany (see section 1.1.5). Laser induced fluorescence (LIF) measurements of atomic excited state hydrogen, nitrogen, and also argon have been conducted using continuous wave (CW) pumped dye lasers to determine velocity profiles and temperatures for comparison purposes with the computational model. LIF was also used to attempt to measure fluctuations in velocity by observing two probe beams simultaneously probing the flow from different angles. Spectroscopic emission measurements were 145 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. also conducted to examine the NH rotational populations and temperatures, electron densities (using the hydrogen Balmer series), and average velocities through Doppler shift measurements. Computational model work determined that constrictor geometry changes could significantly affect plume temperature, but would not alter the overall energy flux. Two processes in the models that were considered most important in building a valid model were determined to be diffusive transport and nonequilibrium recombinatioiL An effort to more closely examine these processes outside of the arcjet environment has begun in order to later improve the arcjet model accuracy. 1.1.10 M a s s a c h u s e tts I n s t i t u t e o f t e c h n o l o g y * • 233.2391 A computational modeling effort to describe the behavior of arcjet operation has been underway at MIT for several years. A Navier-Stokes / McCormick method based fluid dynamic code has been developed that models the arcjet flow and has evolved to contain two fluids and two temperatures, viscosity, heat conduction, ohmic heating, ambipolar diffusion, collisional energy transfer, dissociation, and ionization. The model relaxes from a start condition until the electric potential stabilizes. Data from Institut Filer Raumfahrtsysteme at the University of Stuttgart (Germany) for a 20kW hydrogen arcjet has been used to compare with the computational results. Recently, a change in the code to support 1 kW hydrogen arcjets has taken place and operating conditions similar to that of USC/Phillips Laboratory (see section 1.1.3) and Stanford University (see section 1.1.8) are being run. 146 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. l . L l l T exa s T e c h u n i v e r s i t y 240.24s] Texas Tech University has been the primary place where arcjet materials research has taken place. They have specialized in electrode materials testing and configuratiotL The notion that the best material currently identified for use as an arcjet cathode is 2% thoriated tungsten has been challenged and subsequently confirmed by Texas Tech. Approximately 10 different electrode materials were identified out of a larger group of eligible choices to be tested over a several year period on life and erosion properties. Materials were tested at the 30 kW level, often on nitrogen propellant both in arc simulators and in arcjet configurations. Anodes were also segmented to better understand the arc attachment process and how it contributes to the erosion mechanisms that limit electrode life and ultimately arcjet life. A new arcjet power supply was designed for these tests with the ability to continuously provide voltage vs. current characteristics during the phase change of the power supply ripple. This allowed an electrode life test to continue uninterrupted as significant information was collected on how the electrode's property changes with additional operation over a range of voltages and currents while maintaining a constant average voltage for the life test. Investigation into the effect of the power supply ripple on erosion was also undertaken and found that approximately 10% peak-to-peak ripple is beneficial in elongating electrode life, though increased ripple was detrimental. A qualitative model explained that this was due to the ripple moving the arc away from the cathode tip where the majority of detrimental erosion takes place. Increased ripple caused instability in arcjet operation and could actually increase erosion. 147 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1.1.12 UNIVERSITY OF 1 LUNDIS 246- 249} The University of Illinois' electric propulsion program has explored arcjet operation in the laboratory and modeled it on computer. The primary model configuration is for a 1 kW hydrazine or hydrogen arcjet and is a two-dimensional, Navier-Stokes steady-state solution that assumes no local thermodynamic equilibrium, and includes terms for chemical and thermal nonequilibrium processes. It supports two temperatures and multiple species. Many experiments have been conducted at University of Illinois utilizing electrostatic probes for measuring current density, electron density, electron temperature, and exit plane velocity of ions (with time-of-flight methods). Integration of probes into the wall of the arcjet nozzle have allowed the current densities, electron densities, and electron temperatures to also be measured throughout the arc attachment region providing insight into proper modeling of arc attachment. A development program to create a low-power pulsed arcjet system is also underway at University of Illinois. The intention is to increase the specific impulse of arcjet operation by only operating during very high current pulses. Microsecond pulses operated thousands of times per second on hydrazine or helium indicate that an increase in specific impulse is possible, but perhaps at a substantial cost in life and wear to the electrodes. Continuous operation for extended time periods has yet to be demonstrated. /. 1.13 O h io S ta t e U n iv e r s ity 37, 250-2531 Ohio State University has multiple electric propulsion programs ongoing, though only one program focuses on arcjet research. This cotnputational effort is believed by some to be the most comprehensive modeling effort currently underway, with detailed attention to specific chemical 148 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. nonequilibrium processes that OSU believes significantly affect the outcome of the processes in the arcjet thruster. The arcjet modeling code is only recently being constructed after completing development on many different code modules to handle specific physical processes. The code is a Navier-Stokes based solution that has been transitioned from a one-dimensional code to a two-dimensional code that also encompasses swirl. Physical processes that comprise the model include molecular vibration population mechanisms, rotational population, and electronic excitation. Dissociation and ionization processes are also closely examined while relaxation modes for translation and vibration are also modeled. Many chemical reactions are modeled including rare but extremely fast hydrogen reactions that OSU believes may play a significant role in the dissociation process. The model supports nitrogen, hydrogen, and consequently also ammonia and hydrazine at high and low powers. The geometry is currently designed for high power arcjets, but efforts are also underway to support the operating conditions similar to that of USC/Phillips Laboratory {see section 1.1.3} and Stanford University (see section 1.1.8). / . / . 14 Un iv e r s it y o f M ic h ig a n ts4.2ssi The University of Michigan has just recently entered electric propulsion research and has begun by exploring arcjet plume properties in a new world-class vacuum facility where a diagnostic repertoire is being developed. Arcjet performance on hydrogen has been explored with thrust stand measurements and a range of operating conditions at very low vacuum chamber pressures has been examined. Emission spectroscopy measurements have examined electron densities and temperatures through 149 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. examinations of the hydrogen Balmer series as well as temperatures of the excited state atomic hydrogen. A microwave interferometer was developed at Michigan to measure electron densities in the far field and compare them with the emission spectroscopy measurements. Electron temperatures, pressures, and flow field patters were also examined with this device. 1.1. IS Co r n e l l Un iv e r s it y Cornell University has a program exploring the use of particle-based computational models in particular the Direct Simulation Monte Carlo (DSMC) method. Recently, through collaborations with NASA Lewis and now Stanford University, the DSMC method is being applied to the solution of 1 kW hydrogen arcjet simulations. Initially a solution of the cold flow of hydrogen through the arcjet nozzle was undertaken and compared with Raman spectroscopy data on hydrogen molecules from Stanford (see section 1.1.8). This proof-of-concept work demonstrated the usefulness of applying the DSMC method to the arcjet problem and the potential that DSMC has to handle the highly non-equilibrium conditions of the arcjet flow when the arc is present. The cold flow effort compared molecular number density, rotational temperatiue and population of the first rotational level. Sensitivity to the rotational relaxation rate was observed in the simulation and a strong degree of thermal nonequilibrium was observed even in the cold flow. Determination of the relaxation rate that allowed the model to agree with experiment was reasonable and Cornell feels that this provides a major step towards the modeling of hot flow through the arcjet. Collaborations on the arcjet model continue with Stanford and are also planned with Olin Aerospace and Phillips Laboratory/USC. 150 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. atomic H density (attempts) NH density I < % < z I < C O i S c I I I I % "e > " 5 « C Z I u I I K N k 4 ) S âi s 3 " O I 3 > "c 3 I :I 3 c 3 £ I o Emission Spectroscopy / Microwave Interferomei a electron number density electron temperature excited state temperature excited state velocity vibrational temperature rotational temperature Electrostatic Probes electron number density electron temperature ion drift velocity Laser Induced Fluorescence excited state velocity excited state temperature molecular H2 density (cold flow) molecular H2 temperature (cold flow) molecular H2 pressure (cold flow) Coherant Anti-Raman Stokes i molecular H2 velocity (resitojet) molecular H2 temperature (resistojet) 1 Electrode Research Power Supply - Arcjet Interaction Mass Spectrum Analysis H iuiim Computational Models Navier-Stokes based model DSMC model includes non-equilibrium includes chemistry includes large number of reactions a Figure 1-1: Research institutions and the arcjet research areas that each has participated in. 151 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1.2 D is c u s s io n o n A r c j e t R e s e a r c h Upon examining the sections above (summarized in Figure 1-1), it appears clear that electron density and temperature measurements have been conducted by many different research groups and further work in that area does not appear to be necessary. The six computational models that are being developed all have incorporated data from at least one of these electron density/temperature measurements. Likewise, many excited state temperature measurements have been conducted and the models have often used this temperatime data for calibration even though species specific temperatures are not supported in most of the models (most have one temperature for “heavy” particles and one temperature for electrons). Emission measurements by many groups clearly indicate tiigh levels of nonequilibrium in the arcjet plume questioning the entire concept of an overall species “temperature” in the first place. Vibrational and rotational temperature measurements have been made on NH and N; in ammonia/hydrazine arcjets with a difference of opinion by many of the research groups on whether or not enough levels are in equilibrium to allow construction of a valid Boltzman plot and subsequently an equilibrium temperature measurement. Species temperature and velocity have also been measured in a point-specific manner using laser induced fluorescence. LIF is also an excited state measurement and tfiis brings up the question of the usefulness to the current models of this temperature measurement in a non- equilibrium plasma. The velocity data would appear to accurately describe hydrogen atoms in a hydrogen (or hydrogen/nitrogen) arcjet as similar species (hydrogen and helium) have been observed to have indistinguishable velocities inferring that excited state and ground state hydrogen 152 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. likely have similar velocities. Velocity of the molecular species has not been measured except in the far field plume by mass spectroscopy. What appears to be currently missing is species density. So far, two attempts at atomic hydrogen density measurements have been made and each has questioned whether or not absorption spectroscopy has much of a chance at successfully measuring atomic hydrogen concentrations. Molecular densities have been successfully measured in an arcjet flow with no electricity applied to the arcjet. Measurements o f this type (Raman scattering) will be substantially more difficult in a lit arcjet where a bright emission background will make signal-to-noise a significant problem. Densities of species other than electrons or ions such as nitrogen or NH have not been examined and may be prematiue as most models are attempting to accurately model the simpler case of hydrogen propellant alone before adding nitrogen species and the additional chemistry that goes with it. It would appear that accurate measurements of hydrogen atomic and molecular densities would prove most beneficial to the computational modeling researchers in order to validate mass flux, momentum flux, dissociation fraction, and the significance of losses in the nozzle boundary layer in the computational simulations. Determination of the molecular ground state temperature and the atomic ground state temperature would also be of interest depending upon how similar or different they are with respect to the excited state temperatures that have been previously reported and used in the models. Currently there are two efforts underway to measure the physical parameters of the hydrogen molecules. The Raman scattering effort underway at Stanford has promise that it may overcome the signal-to-noise limitations of a lit arcjet and plans for Coherent Anti-Raman Stokes measurements on arcjets at Aerospace Corporation are on hold, but may soon be performed. 153 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. As of yet, no new plans to examine atomic hydrogen densities have been indicated and two attempts have been made with inconclusive results. It is this area that needs to be addressed in order for the computer models to be validated and is necessary for continued progress in understanding the physical processes that take place in an arcjet thruster. Recent advances in flame diagnostics indicate that methods used for observing concentrations of atomic hydrogen in flames using two photon or multiphoton laser induced fluorescence may be applicable to arcjet plumes and it is the examination of these techniques and their application to the measurement of ground state hydrogen atom concentrations that is proposed. 1.3 L a s e r D e t e c t i o n o f A t o m i c H y d r o g e n Observing atomic hydrogen during its lifetime in a non-intrusive manner often involves optical detection of the emission firom the depopulation of excited electronic states or the measurement of photon absorption through a population of atomic hydrogen. Inducing the fluorescence with laser light tuned to a specific transition frequency (laser induced fluorescence or LEF) can allow information on the species participating in the electronic transition to be inferred. Usually, the excited states of hydrogen are probed as the excited state transitions are more accessible to laser equipment currently available and can be detected in the visible wavelength range. In a non-equilibrium plasma region such as the arcjet plume, the excited state population of atomic hydrogen may not fully represent the properties of the ground state hydrogen atoms and it is desirable to identify a direct method of probing the ground state of hydrogen. Additionally, since a small unknown percentage of hydrogen is found in the excited state, density measurements of atomic hydrogen are best conducted on the ground state population. 154 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. In the flame environment, atomic hydrogen plays an important role in the chemistry of most combustion processes. Measuring ground state hydrogen profiles in premixed flames is considered critically important in understanding how to best model the combustion processes of flame chemistry, and the development of laser diagnostics to non-intrusively probe ground state hydrogen atoms has been explored in depth during the last decade in order to facilitate these measurements. Research in this field has centered on tfiree excitation schemes of ground state hydrogen: two photon direct excitation, three photon direct excitation, and two-step three photon excitation. The latter scheme was developed at Sandia National laboratory and work on the other two schemes has also been examined in detail at Sandia as well as at numerous other institutions. The following sections describe work performed at some of these research facilities around the world that have helped develop these laser-based diagnostic techniques that measure densities of atomic hydrogen in flame environments. 1.3.1 S a n d ia N a tio n a l L a b o r a to r y ^^6-2611 J. E. M. Goldsmith has led a large effort at Sandia National laboratories to investigate laser diagnostics and their applicability to combustion problems. The researchers at Sandia examined the uses of multiple-photon stimulated emission to promote ground state radicals to an excited state whereby emission in the visual wavelength region could be detected. Problems inherent in the technique were investigated and included the dependence of the emission signal on tfiree parameters: laser power, ambient gas pressure, and tfie spectral stiape of the stimulated emission. Multiple photon excitation schemes were found to have increased sensitivity to each of these parameters as compared with standard single photon laser induced fluorescence. 155 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Taking this diagnostic approach to the hydrogen atom was demonstrated using a two-step method where a two photon transition was used to excite the ground state of hydrogen to the first excited state (n=2) and then a resonant detection scheme, requiring a second laser, excited some of those placed in the first excited state to the second excited level (n=3) via a single photon transition. The resulting decay from n=3 to n=2 is observed. The n=l to n=2 transition requires a single photon at a wavelength of 121.5 nm or a two- photon transition at 243 nm (the absorption cross-section for a two photon transition is substantially smaller than for a single photon transition and will be discussed later). The latter wavelength is more desirable for two reasons. The 243 nm light is much more easily created with the laser technology available today (including doubling/tripling crystals and pressure cells) and wavelengths below 2(X) nm are easily absorbed by the atmosphere and require a vacuum path for the entire path of the beam. To complete the two-step method mentioned above, an additional 656 nm beam promoted those atoms tfiat were previously excited (based upon the cross-section probability) to n=3. Subsequent detection of 656 nm emission is observed. Comparison of this technique with direct two-photon and three-photon schemes (discussed in sections 2.1.3 and 2.1.4) was conducted by researchers at Sandia and found to provide more accurate results for rich hydrocarbon flames, though almost identical results for most simple flames. The two step technique may be more difficult to implement than the other techniques as it • requires two lasers (though a variation using different transitions can make be accomplished with only one laser) making the experiment more complex and costly. One of the “side effects” of the multiphoton direct excitation techniques (see sections 2.1.3 and 2.1.4) in hydrogen is that they directly promote ground state electrons to the second excited level 156 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. thereby causing a potential population inversion between the first excited state and the second excited state. This takes place only along the line of irradiation by the ultraviolet laser beam (typical for irradiation of ground state hydrogen atoms) and can lead to the creation of an additional laser beam (in the visible for hydrogen when, for example, Balmer alpha radiation is initiated) along the path of the initial beam. This reabsorption of emitted photons can cause a problem if a measurement is attempting to create an analog between light collected and the density of emitters and must either be accounted for or avoided. Researchers at Sandia were the first to examine this optically excited stimulated emission (sometimes called “amplified spontaneous emission" or ASE) and its relationship to the multiphoton fluorescence diagnostic. One of the important observations was that while the fluorescence due to the excitation followed a laser power-squared behavior, the ASE fluorescence went as laser energy to a higher exponential power meaning that it could dominate at higher laser energies and drop to an insignificant process at lower laser energies. Using the ASE as a diagnostic itself was examined, but its line-of-sight nature and the requirement of transition saturation before a linear relation between species density and signal is observed led to difficulties in application as a diagnostic. In order to use the multiphoton diagnostic technique as a measure of atomic hydrogen density a correlation between the number of emitters and the measured fluorescence is required. A process which can prematurely depopulate atoms excited by the diagnostic is collisional quenching due to even a moderately high pressure environment. Sandia has looked at the time-resolved fluorescence decay of the atomic hydrogen in low pressure flames (20 mtorr or -3 Pa) to examine the relationship of presstire to quenching for this diagnostic technique and has characterized the relationship between the fluorescence decay times and the expected attenuation of the emitted 157 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. fluorescence. Changes in pressure, and correspondingly in quenching, can be determined through observation of changes in fluorescence decay times of the LIF. Relative hydrogen density measurements were compared with flame models using these techniques and the results were found to agree well. No calibrated atomic hydrogen density measurements were made at Sandia to this author’s knowledge. Additional experiments involving clever use of laser equipment when performing multiphoton LIF and experimental setups that accomplish multiple tasks were also investigated at Sandia. 1.3.2 INSTIW T FÜR PHYSIKAUSCHE CHEMIE DER VERBRENNVNG (GERMANY) 26-291 Detailed examinations of quenching, excitation linewidth, and fluorescence calibration for two-photon LIF diagnostics on atomic hydrogen in flames have taken place at the Institute for the Physical Chemistry of Combustion at the University of Stuttgart. Direct two-photon laser induced fluorescence at 205 nm was implemented to promote ground state hydrogen atoms to the n=3 excited state where n=3 to n=2 (Balmer a) fluorescence was observed. Key to the work performed at Stuttgart was detailed calibration of the fluorescence using a calibration cell containing atomic hydrogen created from a microwave discharge. Hydrogen densities in the cell were determined using a chemical gas titration technique that is discussed later in greater detail (see section 2.3). Additionally, the technique was successfully demonstrated near a heated filament where pressures ranged from 1 to 100 mbar (100 to 10,000 Pa) and the hydrogen source came from a 5% CRi chamber environment. Clenching effects were calculated in pressure ranges above 10 mbar and determined from fluorescence lifetime decay rates at lower pressures as described earlier (see section 1.3.1). 158 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1.3.3 O h io S t a te Un iv e r s it y OSU also has done several multiphoton LIF experiments similar to the groups discussed previously. Of interest in their publications is additional attention to detail involving chemical titration calibration of two-photon LIE and corrections for coUisional quenching. Researchers at OSU thoroughly describe the process of determining atomic hydrogen density in a calibration cell using titration of small amounts of NO: to chemically react with hydrogen atoms present in the cell (see section 2.3). Additionally, they have examined in detail the fine structure of the r^3 state of atomic hydrogen in order to determine the unquenched radiative lifetime of this excited state when populated by a two-photon excitation process. This lifetime is important in determining the correction factor applied to a quenched signal through measurement of fluorescence decay. 1.3.4 K yu sh u u n i v e r s i t y (Ja p a n ) 262.2031 The Department of Energy Conversion at Kyushu University has spent time examining laser diagnostics for use in high temperature plasmas. While these plasma environments are substantially different than the plasma, environments found in electric propulsion thrusters such as arcjets, the use of laser diagnostics in these environments is of interest. The group at Kyushu has endeavored to compare various laser diagnostics for use in these high temperature plasmas and has closely compared four different two-photon excitation schemes. Studies from Kyushu found that unless hydrogen densities are very low (<10“ cm'^) losses in detecting vacuum UV emission require the use of fluorescence emission in the visible to detect hydrogen atoms and this constrains the choice of excitation level to either the n=3 or n=4 excited state. Of these options the Kyushu group determined that two-photon excitation into the n=3 level and the subsequent observation at the Balmer a wavelengths produced the largest number of 159 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. fluorescence photons per incident laser photon. This also was the best choice if molecular dissociation due to the incident laser was to be avoided and if incident intensities were to be kept low (to avoid ASE, for example, as described in section 1.3.1). In addition, two photon excitation from n=l to n=3 also appeared to be the least sensitive to coUisional quenching of the four techniques that were examined. Due to aU of the above advantages, it appeared that for an environment somewhat similar to that in electric propulsion plasmas, the two-photon to n=3 scheme appeared to be the technique of choice. 160 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2 References Appendix ’ W. W. Smith, “Low Power Hydrazine Arcjet Qualification,” Paper IEPC-9I-148, 22“* International Electric Propulsion Conference, Viareggio, Italy, 14-17 Oct 1991. ■ C. E. Vaughan, and R. J. Cassady, "An Updated Assessment of Electric Propulsion Technology for Near-Earth Space Missions,” Paper AIAA-92-3202, 28“ ^ Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. ^ D. H. Manzella and M. A. Cappelli, “Vacuum Ultraviolet Absorption in a Hydrogen Arcjet,” Paper AIAA-92-3564, 23"^ Plasmadynamics and Lasers Conference, Nashville, Teimessee, 6-8 July, 1992. ■ * J. E. Pollard, “Arcjet diagnostics by XUV Absorption Spectroscopy,” Paper AIAA-92- 2966, 23"* Plasmadynamics and Lasers Conference, Nashville, Tennessee, 6-8 July, 1992. ^ D. Keefer, D. Burtner, T. Moeller, and R. Rhodes, “Multiplexed Laser Induced Fluorescence and Non-Equilibrium Processes in Arcjets,” Paper AIAA-94-2656, 25'*' Plasmadynamics and Lasers Conference, Colorado Springs, Colorado, 20-23 July, 1994. ® G. W. Butler, I. D. Boyd, and M. A Cappelli, “Non-Equilibrium Flow Phenomena in Low Power Hydrogen Arcjets,” Paper AJAA-95-2819, 31“ Joint Propulsion Conference, San Diego, California, 10-12 July, 1995. V. Babu, S. M. Aithal, and V. V. Subramaniam, “Propellant Internal Mode Dis-equilibrium and Frozen flow Losses in arcjets,” Paper AIAA-94-2655, 25'*’ Plasmadynamics and Lasers Conference, Colorado Springs, Colorado, 20-23 July, 1994. * S. Miller and M. Martinez-Sanchez, "Nonequilibrium Numerical Simulation of Radiation-Cooled Arcjet Thrusters,” Paper IEPC-93-218, 23"* International Electric Propulsion Conference, Seattle, Washington, 13-16 September, 1993. ’ T. W. Megli, H. Krier, R. L. Burton, and A. E. Mertogul, "Two Temperature Modeling of NTH, Arcjets,” Paper AIAA- 94-2413, 25'*’ Plasma-dynamics and Lasers Conference, Colorado Springs, Colorado, 20- 23 July, 1994. I. D. Boyd, M. A Cappelli, and D. R. Beattie, “Monte Carlo And Experimental Studies Of Nozzle Flow In A Low-Power Hydrogen,” Paper AlAA-93-2529, 29'*’ Joint Propulsion Conference, Monterey, C A 28- 30 June, 1993. * * D. R. Beattie and M. A Cappelli, “Molecular Hydrogen Raman Scattering in a 161 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Low Power Arcjet Thruster,” Paper AIAA- 92-3566, 28“* Joint Propulsion Conference, Nashville, Tennessee, 6-8 July, 1992. D. R. Beattie, and M. A. Cappelli, “Raman Scattering Measurements of Molecular Hydrogen in an Arcjet Thruster Plume," Paper AIAA-95-1956, 26* Plasmadynamics and Lasers Conference, San Diego, CA, June, 1995. D. A. Erwin, G. C. Pham-Van-Diep, and W. D. Deininger, “Laser-induced Fluorescence Measurements of Flow Velocity in High-Power Arcjet Thruster Plumes,” AIAA Journal 29, 1298 (1991). J. G. Liebeskind, R. K. Hanson, and M. A. Cappelli, “Laser-induced Fluorescence Diagnostic for Temperature and Velocity Measurements in a Hydrogen Arcjet Plume," Applied Optics 32, 6117 (1993). A. C. Eckbreth, Laser Diagnostics for Combustion Temperature and Species, 1 “ ed., ABACUS Press, Kent, 1988. W. Demtroder, Laser Spectroscopy: Basic Concepts and Instrumentation, 3 "* Printing, Springer-Verlag, Berlin, 1988. J. E. M. Goldsmith, ‘Two-Photon-Excited Stimulated Emission From Atomic Hydrogen In Flames,” J. Opt. Soc. Am. B 6, 1979, 1989. J. E. M. Goldsmith and N. M. Laurendeau, “Single-Laser Two-Step Fluorescence Detection of Atomic Hydrogen in Flames,” Optics Letters 15, 10, 576-578, 1990. J. E. M. Goldsmith, “Multiphoton-Excited Fluorescence Measurements O f Atomic Hydrogen In Low-Pressure Flames," Proceedings of the 22“* Symposium (International) on Combustion, Combustion Institute, 1403, 1988. W. H. Press, B. P. Flannery, S. A. Teukolsky, and W. T. Vetterling, Numerical Recipes: The Art o f Scientific Computing, 1 * * Ed., Cambridge Univ. Press, Cambridge, 1986. D.J. Bamford, L.E. Jusinski, and W.K. Bischel, “Absolute two-photon absorption and three-photon ionization cross sections for atomic oxygen,” Phys. Rev A 34, 185-198 (1986). “ N. Georgiev, K. Nyholm, R. Fritzon, and M. Alden, “Developments of the amplified stimulated emission technique for spatially resolved species detection in flames.” Optics Comm. 108.71-76(1994). ^ M.S. Brown and J.B. Jeffries, “Measurement of atomic concentrations in reacting flows through use of stimulated gain or loss,” Applied Optics 34, 1127 (1995). ^ P.V. Storm and M.A. Cappelli, “High Spectral Resolution Emission Study of a Low Power Hydrogen Arcjet Plume,” Paper AIAA 95-1960, 26* Plasmadynamics and Lasers Conference, San Diego, California, 19-22 June, 1995. ^ C.R. Vidal, J. Cooper, and E.W. Smith, “Hydrogen Stark-broadening tables,” Astrophys. J. Suppl. No. 214, vol. 25. 37- 136 (1973). “ U. Meier, K. Kohse-Hoinghaus, L. Schafer, and C.-P. KJages, ‘Two-photon excited LIF determination of H-atom concentrations near a heated filament in a 162 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. low-pressure H, environment,” Appl. Opt. 29,4993, 1990.' J. Bittner, K. Kohse-Hoinghaus, U. Meier, S. Kelm, and T.H. Just. “Determination of Absolute H Atom Concentrations in Low- Pressure Flames by Two-Photon Laser- Excited Fluorescence," Combustion and Flame 2L 41-50, 1988. U. Meier, J. Bittner, K. Kohse-Hoinghaus, and Th. Just, “Discussion of Two-Photon Laser-Excited Fluorescence As a Method For Quantitative Detection of Oxygen Atoms in Flames," 22“* Symposium (International) on Combustion, The Combustion Institute, 1887-1896, 1988. U. Meier, K. Kohse-Hoinghaus, and Th. Just, “H And O Atom Detection For Combustion Applications: Study Of 1986) Quenching And Laser Photolysis Effects,” Chem. Phys. Lett. 126. 567, 1986. “ A. D. Tserepi, J. R. Dunlop, B. L. Preppemau, and T. A. Miller, “Absolute H- atom Concentration Profiles in Continuous and Pulsed RF Discharges,” J. Appl. Phys. 72, No. 7, 2638(1992). F.M. Curran and T.W. Haag, “An Extended Life and Performance Test of a Low Power Arcjet,” Paper AIAA-88-3106, 24th Joint Propulsion Conference, New York, New York, 1988. B.L Preppemau, D. A. Dolson, R.A. Gottscho, and T. A. Miller, “Temporally Resolved Laser Diagnostic Measurements Of Atomic Hydrogen Concentrations In RF Plasma Discharges," Plasma Chem. and Plasma Proc. 9, 157 (1989). G.W. Buder, A.E. Kull, and D.Q. King, “Single fluid simulations of low power hydrogen arcjets,” 30th Joint Propulsion Conference, 27-29, Indianapolis, Indiana, June, 1994. ” M. W. Crofton, R. P.Welle, S. W. Janson, and R. B. Cohen, , “Rotational and Vibrational Temperatures in the Plume of a 1 kW Ammonia Arcjet,” AIAA Paper 91- 1491, June 1991. I. J. Wysong, J. A. Pobst, I. D. Boyd, “Comparisons of Hydrogen Atom Measurements in an Arcjet Plume with DSMC Predictions,” Paper AIAA-96-3185, 32“^ Joint Propulsion Conference, Lake Buena Vista, FL, July 1996. T. W. Megli, H. Krier, R. L. Burton, “A Plasmadynamics Model for Nonequüibrium Processes in N^/H: Arcjets,” Paper AIAA- 95-1961, 26* Plasmadynamics and Lasers Conference, San Diego, CA, June 1995. ” S. M. Aithal, V. V.Subramaniam, “Effects of Arc Attachment on Arcjet Flows,” Paper AIAA-96-3295, 32“ '’ Joint Propulsion Conference, Lake Buena Vista, FL, July 1996. J. G. Liebeskind, R. K. Hanson, M. A. Cappelli, “LIF Measurements of Species Velocities in an Arcjet Plume," Paper lEPC- 93-131, 23"’ International Electric Propulsion Conference, Seattle, WA, September 1993. E. J. Pencil, J. M. Sankovic, C. J. Sarmiento, and J. A. Hamley, “Dependence of Hydrogen Arcjet Operation on Electrode Geometry,” Paper AIAA-92-3530, 28* Joint Propulsion Conference, Nashville, TN, July 1992. 163 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 0 J. M. Sankovic, F. M. Curran, “A Low- Erosion Starting Technique For High- Performance Arcjets,” Paper AIAA-94-2867, AIAA/AS ME/S AE/ASEE Joint Propulsion Conference, 30*, Indianapolis, IN, 27-29 Jun 1994. 4 1 J. M. Sankovic, “Ultra-Low-Power Arcjet Thruster Performance,” CPIA-PUB-602- VOL-V, 1993 JANNAF Propulsion Meeting, Monterey, CA, 15-19 Nov 1993. 4 2 F. Curran, L. Caveny, “Hydrogen Arcjet Technology Status,” Paper IEPC-93-2I5, 23"* International Electric Propulsion Conference, Seattle, WA, 13-18 Sept 1993. 4 3 V. K. Rawlin, L. R. Pinero, J. A. Hamley, “Simplified Power Processing For Inert Gas Ion Thrusters,” Paper AIAA-93-2397, 29* Joint Propulsion Conference, Monterey, CA, 28-30 Jun 1993. 4 4 J . A. Hamley, J. M. Sankovic, “A Soft- Start Circuit For Arcjet Ignition,” Paper AIAA-93-2396, 29* Joint Propulsion Conference, Monterey, CA, 28-30 Jun 1993. 45 E. J. Pencil, C. J. Sarmiento, D. A. Lichtin, J. W. Palchefsky, “Low Power Arcjet System Spacecraft Impacts,” Paper AIAA- 93-2392, 29* Joint Propulsion Conference, Monterey, CA, 28-30 June 1993. 4 6 F. M. Curran, J. R Brophy, G. L. Bennett, “The NASA Electric Propulsion Program,” Paper AIAA-93-1935, 29* Joint propulsion Conference, Monterey, CA, 28-30 Jun 1993. 47 T. W. Haag, “Recent Testing Of 30 kW Hydrogen Arcjet Thrusters,” Paper AIAA- 93-1902, 29* Joint Propulsion Conference, Monterey, CA, 28-30 June 1993. 4 8 D. Bems, J. Sankovic, C. Sarmiento, “Investigation Of A Subsonic-Arc- Attachment Thruster Using Segmented Anodes,” Paper AIAA-93-1899, 29* Joint Prqjulsion Conference, Monterey, CA, 28- 30 June 1993. 49 J. Sankovic, D. Bems, “Performance Of A Low-Power Subsonic-Arc-Attachment Arcjet Thruster,” Paper AIAA-93-1898, 29* Joint propulsion Conference, Monterey, CA, 28- 30 June 1993. 5 0 F. M. Curran, T. W. Haag, “Extended life and performance test of a low-power arcjet,” Journal of Spacecraft and Rockets, 29, 444- 52, July / August 1992. 51W. E. Morren, P. J. Lichon, “Low-Power Arcjet Test Facility Impacts,” Paper AIAA- 92-3532, 28* Joint Propulsion Conference, NashvUle, TN, 6-8 Jul 1992. 5 2 J. A. Hamley, L. R. P*inero, G. M. Hill, “The 10 kW Power Electronics For Hydrogen Arcjets,” CPIA-PUB-580-VOL-1, 1992 JANNAF Propulsion Meeting, Indianapolis, IN, 24-27 Feb 1992. 53 J. M. Sankovic, F. M. Curran, C. A. Larson, “Effects Of Anode Material On Arcjet Performance,” CPIA-PUB-580-VOL- 1, JANNAF Propulsion Meeting, Indianapolis, IN, 24-27 Feb 1992. 5 4 J. S. Sovey, J. A. Hamley, M. J. Patterson, V. K. Rawlin, R.M. Myers, “The Evolutionary Development Of High Specific Impulse Electric Thrust,” AIAA-92-1556, 1992. 55 J. M. Sankovic, J. A. Hamley, T. W. Haag, C. J. Sarmiento, F. Curran, “Hydrogen Arcjet Technology,” Paper EEP*C-91-018, 22“* International Electric Propulsion Conference, Viareggio, Italy, 14-17 Oct 1991. 5 6 J. M.Sankovic, F. M. Curran, “Arcjet Thermal Characteristics,” Paper AIAA-91- 2456, 27* Joint Propulsion Conference, Sacramento, CA, 24-26 Jun 1991. 164 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 5 7 V. te Rawlin, G. A. Majcher, “Mass Comparisons Of Electric Propulsion Systems For NSSK Of Geosynchronous spacecraft,” Paper AIAA-91-2347, Z?* Joint Propulsion Coifterence, Sacramento, CA, 24-26 Jun 1991. 5 8 W. E. Morren, F. M. Curran, “Preliminary Performance And Life Evaluations Of A 2- kW Arcjet,” Paper AlAA-91-2228, 59 F. M. Curran, S. R. Bullock, T. W. Haag, C. J. Sarmiento, J. Sankovic, “Malium Power Hydrogen Arcjet Operation,” Paper AIAA-91-2227, 27* Joint Propulsion Conference, Sacramento, CA, 24- 26 Jun 1991. 60 T. w. Haag, F. M. Curran, “High Power Hydrogen Arcjet Performance," Paper AIAA-91-2226, 27* Joint Propulsion Conference, Sacramento, CA, 24-26 Jun 1991. 61 J. A. Hamley, G. M. Hill, “Power Electronics For Low Power Arcjets,” Paper AIAA-91-1991, 27* Joint Propulsion Conference, Sacramento, CA, 24-26 Jun 1991. 6: F. M. Curran, C. J. Sarmiento, B. W. Birkner, J. Kwasny, “Arcjet Nozzle Area Ratio Effects,” CPIA-PUB-550-VOL-2, 1990 JANNAF Propulsion Meeting, Anaheim, CA, 3-5 Oct 1990. 6 3 J. M. Sankovic, “Investigation Of The Arcjet Near Field Plume Using Electrostatic Probes,” CP1A-PUB-550-VOL-2, 1990 JANNAF Propulsion Meeting, Anaheim, CA, 3-5 Oct 1990. 6tD. H. Manzella, F. M. Curran, R. M. Myers, D. M. Zube, “Preliminary Plume Characteristics Of An Arcjet Thruster,” Paper AIAA-90-2645, 21“ International Electric Propulsion Conference, Orlando, FL, 18-20 Jul 1990. 6 5 J. A. Hamley, “Arcjet Load Characteristics,” Paper AIAA-90-2579, 21“ International Electric Propulsion Conference, Orlando, FL, 18-20 Jul 1990. 6 6 F. M. Curran, C. J. Sarmiento, “Low Power Arcjet Performance,” Paper AIAA- 90-2578. 21“ International Electric Propulsion Conference, Orlando, FL, 18-20 Jul 1990. 67J. S. Sovey, D. J. Pidgeon, “Advanced propulsion For LEO And GEO Platforms,” Paper AlAA-90-2551, 21“ International Electric Propulsion Conference, Orlando, FL, 18-20 Jul 1990. 6 8 R. P. Gruber, R. W. Gott, T. W. Haag, “5-Kw Arcjet Power Electronics,” Paper AIAA-89-2725, 25* Joint Propulsion Conference, Monterey, CA, 10-12 Jul 1989. 6 9 L. M. Camey, J. M. Sankovic, “The Effects Of Arcjet Operating Condition And Constrictor Geometry On The Plasma Plume,” Paper AIAA-89-2723, 25* Joint Propulsion Conference, Monterey, CA, 10- 12 Jul 1989. 7 0 F. M. Curran, D. H. Manzella, “The Effect Of Electrode Configuration On Arcjet Performance,” Technical Memorandum NASA-TM-102346, 25* Joint Propulsion Conference, Monterey, CA, 10-12 Jul 1989. 71F. M. Curran, A. J. Sovie, T. W. Haag, “Arcjet Nozzle Design Impacts,” CPIA- PUB-515-VOL-1, 1989 JANNAF Propulsion Meeting, Cleveland, OH, 23-25 May 1989. 73 F. M. Curran, T. W. Haag, J. F. Raquet, “Arcjet Cathode Phenomena,” CPIA-PUB- 515-VOL-l, 1989 JANNAF Propulsion Meeting, Cleveland, OH, 23-25 May 1989. 73 M. J. Patterson, F. M. Curran, “Electric Propulsion Options For 10 kW Class Earth- 165 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Space Missions,” CPIA-PUB-515-VOL-l, 1989 JANNAF Propulsion Meeting, Cleveland, OH, 23-25 May 1989. 7 4 D. C. Byers, “Advanced Onboard Propulsion Benefits And Status,” Technical Memorandum NASA-TM-103174, Symposium on Space Commercialization: Role of Developing Countries, Nashville, TN, 5-10 Mar 19897 7 5 “Comprehensive Report Of Aero Propulsion, Space Propulsion, Space Power, And Space Science Applications Of The Lewis Research Center,” Annual Report, NASA-TM-100925, Jan, 1988. 7 6 F. M. Curran, T. L. Hardy, T. W. Haag, “A Low Power Arcjet Cyclic Life Test,” CPIA-PUB^80-VOL-1, 1987 JANNAF Propulsion Meeting, San Diego, C A 15-17 Dec 1987 7 7 J. S. Sovey, L. M Zana,. S. C. Knowles, “Electromagnetic Emission Experiences Using Electric Propulsion Systems - A Survey,” Paper AIAA-87-2028, 23"' Joint Propulsion Conference, San Diego, C A 29 Jun-2 Jul, 1987. 7 8 C. J. Sarmiento, R. P. Gruber, “Low Power A cjet Thruster Pulse Ignition,” Paper AIAA-87-1951, 23"' Joint Propulsion Conference, San Diego, CA, 29 Jun-2 Jul, 1987. 7 9 L. M. Zana, “Langmuir Probe Surveys Of An Arcjet Exhaust,” Paper AIAA-87-1950, 23"' Joint Propulsion Conference, San Diego, CA, 29 Jun-2 Jul, 1987. 80 T. L. Hardy, F. M. Curran, “Low Power DC A cjet Operation With Hydrogen/Nitrogen/Ammonia Mixtures,” Paper AIAA-87-1948, 23"' Joint Propulsion Conference, San Diego, C A 29 Jun-2 Jul, 1987. 81 J. R. Stone, E. S. Huston, 'The NASA/USAF Arcjet Research And Technology Program,” Paper AAA-87- 1946, 23"' Joint Propulsion Conference, San Diego, CA, 29 Jun-2 Jul, 1987. 82 T. W. Haag, F. M. Curran, "Acjei Starting Reliability: A Multistart Test On Hydrogen/Nitrogen Mixtures,” Paper AAA- 87-1061. 19"* International Electric Prc^ulsion Conference, Colorado Springs, CO, 11-13 May 1987. 8 3 F. M. Curran, T. W. Haag, “Acjet Conçonent Conditions Through A Multistart Test,” Paper AAA -87-1060, 19'" International Electric Propulsion Conference, Colorado Springs, CO, 11-13 May 1987. mT. L. Hardy, V. K. Rawlin, M. J. Patterson, “Electric Propulsion Options For The SP-100 Reference Mission,” Paper NASA-TM-88918,4* Symposium on Space Nuclear Power Systems, Abuquerque, NM, 12-16 Jan 1987. 8 5 J. R. Stone, “NASA Electrothermal Auxiliary Propulsion Technology,” Paper A AA-86-1703, 22“ " Joint Propulsion Conference, Huntsville, AL, 16-18 June 1986. 8 6 F. M. Curran, S. Nakanishi, “Low Power DC A cjet Operation With Hydrogen/Nitrogen Propellant Mixtures,” Paper AIAA-86-1505, 22“ " Joint Propulsion Conference, Huntsville, AL, 16-18 June 1986. 8 7 James S. Sovey, Terry L. Hardy, Matthew Englehart, “A Bibliography Of Electrothermal Thruster Technology, 1984,” Bibliography NASA-TM-86998, 1986. 8 8 T. L. Hardy, F. M. Curran, “Hollow Cathodes In High Pressure A c Discharges,” Paper AAA-85-2035, 18'" International 1 6 6 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Electric Propulsion Conference, Alexandria, VA, 30 S ep-2 Oct 1985. 8 9 S. Nakanishi, “Experimental Performance Of A 1-Kilowatt Arcjet Thruster,” Paper AIAA-85-2033, 18* International Electric Propulsion Conference, Alexandria, VA, 30 Sep - 2 Oct 1985. 9 0 T. L. Hardy, “Electrode Erosion In Arc Discharges At Atmospheric Pressure," Paper AIAA-85-2018, 18* International Electric Propulsion Conference, Alexandria, VA, 30 Sep - 2 Oct 1985. 91F. M. Curran, “An Experimental Study Of Energy Loss Mechanisms And Efficiency Considerations In The Low Power DC Arcjet,” Paper AIAA-85-2017, 18* International Electric Propulsion Conference, Alexandria, VA, 30 Sep - 2 Oct 1985. 9 2 S. Y. Wang, P. J. Staiger, “Space-Based Radar Orbit Transfer,” Paper AIAA-85- 1477,21” Joint Propulsion Conference, Monterey, CA, 8-10 July 1985. 9 3 W. A. Hoskins, G. W. Butler, A. E. Kull, “A Comparison Of Regenerative And Conventional Arcjet Performance,” Paper AIAA-94-3124, 30* Joint Propulsion Conference, Indianapolis, IN, 27-29 Jun 1994. « C . H. McLean, P. G. Lichon, J. Sankovic, “Life Demonstration Of A 600-Second Mission Average Arcjet,” Paper AIAA-94- 2866, 30* Joint Propulsion Conference, Indianapolis, IN, 27-29 Jun 1994. 9 5 W. Hoskins, A. Kull, W. Nesser, “Measurement of Energy Deposition Modes in an Intermediate Power Hydrogen Arcjet,” Paper IEPC-93-216, 23^" International Electric Propulsion Conference, Seattle, WA, 13-18 Sept 1993. 9 6 P. Lichon, J. Sankovic, “Development and Demonstration of a 600 Second Mission Average Arcjet,” Paper EEPC-93-087, 23"' International Electric Propulsion Conference, Seattle, WA, 13-18 Sept 1993. 9 7 R. Aadland, C. Vaughan, W. Hoskins, R. Kay, “Achieving Reliable, Repeatable Starts of a 26 kW Arcjet,” Paper ŒPC-93-049, 23"* International Electric FYopulsion Conference, Seattle, WA, 13-18 Sept 1993. 9 8 C. E. Vaughan, R. J. Cassady, J. R. Fisher, “Design Fabrication and Test of a 26 kW Arcjet and Power Conditioning Unit,” Paper lEPC-93-048, 23* International Electric Propulsion Conference, Seattle, WA, 13-18 Sept 1993. 9 9 R. D. Smith, S. E. Yano, K. Armbruster, C. R. Roberts, D. A Lichtin, J. W. Beck, “Flight Qualification of a 1.8 kW Hydrazine Arcjet System," Paper IEPC-93-007, 23"^ International Electric Propulsion Conference, Seattle, WA, 13-18 Sept 1993. 100C. E. Vaughan, J. P. Morris, “Propellant Feed Subsystem For A 26 kW Flight Arcjet Propulsion System,” Paper AIAA-93- 2400,29* Joint Propulsion Conference, Monterey, CA, 28-30 Jun 1993. 101R. S. Aadland, C. E. Vaughan, R. J. Cassady, R. J. Kay, “Integrated Mission Simulation Of A 26 kW Flight Arcjet Propulsion System,” Paper AIAA-93-2395, 29* Joint Propulsion Conference, Monterey, CA, 28-30 Jun 1993. 102 P. T. Skelly, J. R. Fisher, C. M. Golden, “Power Conditioning Unit For Low-Power Arcjet Flight Application,” Paper AIAA-92- 3529, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 1 0 3 W. A. Hoskins, A E. Kull, G. W. Butler, “Measurement Of Population And 167 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Temperature ProGles In An Arcjet Plume," Paper AIAA-92-3240, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. ic m G. W. Butler, D. Q. King, “Single And Two Fluid Simulations Of Arcjet Performance,” Paper AIAA-92-3104, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. !osC. M. Golden, J. R. Fisher, R. J. Kay, “Power Conditioning Unit For 30 kWe Class Arcjet Flight Experiment,” CPlA-PUB-580- VQL-l, 1992 JANNAF Propulsion Meeting, Indianapolis, IN, 24-27 Feb 1992. 106 R. J. Cassady, C. E. Vaughan, R. S. Aadland, “A 26 kW Arcjet System For On- Orbit Demonstration,” CPlA-PUB-580- VOL-1, 1992 JANNAF Propulsion Meeting, Indianapolis, IN, 24-27 Feb 1992. 1 0 7 Rocket Research Corp. Redmond WA, “H2 Arcjet Performance Mapping Program,” Final Report, Mar 91-Jan 92 NASA-CR- 191073. 108 R. J. Cassady, P. G. Lichon, D. Q. King, “Arcjet Endurance Test Program,” Final Report, Dec 88-Sep 90 AL-TR-90-069, Contract F04611-89-C-0013. 1 0 9 S. C. Knowles, S. E. Yano, "Engineering Model Low Power Arcjet System Development," CPlA-PUB-515-VOL-l, 1 1 01989 JANNAF Propulsion Meeting, Cleveland, OH, 23-25 May 1989. 1 1 1 S. E. Yano, S. C. Knowles, “Simulated Flight Qualification Testing Of An Engineering Model Arcjet System,” CPIA- PUB-515-VOL-l, 1989 JANNAF Propulsion Meeting, Cleveland, OH, 23-25 May 1989. 112 K. Armbruster, R. J. Cassady, R. T. Féconda, “A Pegasus-Derived Electric Propulsion Stage,” CPlA-PUB-515-VOL-l, 1989 JANNAF Propulsion Meeting, Cleveland, OH, 23-25 May 1989. 1 1 3 R. J. Cassady, “An Analysis Of Orbit Maneuvering Capabilities Using Arcjet Propulsion,” Paper AIAA-88-2832,24* Joint Propulsion Conference, Boston, MA, 11-13 Jul 1988. imW. W. Smith, “Electrothermal Thruster Technology Development," CPlA-PUB-480- VOL-1, 1987 JANNAF FYopulsion Meeting, San Diego, CA, 15-17 Dec 1987. 1 1 5 Rocket Research Corp. Redmond WA, “Arcjet Thruster Research And Technology. Phase 1: Executive Summary," Technical Report NASA-CR-182106 Aug, 1987 3 IP, Contract NAS3-24631. 116 M. A. Simon, S. C. Knowles, F. M. Curran, T. L. Hardy, “Low Power Arcjet Life Issues,” Paper AIAA-87-1059, 19* International Electric Propulsion Conference, Colorado Springs, CO, 11-13 May 1987. 1 1 7 S. C. Knowles, W. W. Smith, F. M. Curran, T. W. Haag, “Performance Characterization Of A Low Power Hydrazine Arcjet," Paper AlAA-87-1057, 19* International Electric Propulsion Conference, Colorado Springs, CO, 11-13 May 1987. 118 R. J. Cassady, M. S. Hiatt, “Thrust Measurement Of A 30 kW Arcjet,” 1986 JANNAF Propulsion Meeting, New Orleans, Louisiana, 26-28 August 1986. 1 1 9 S. C. Knowles, W. W. Smith, S.l. Chun, R.T. Féconda, “Low Power Hydrazine Arcjets A System Description For Near- Term Application,” 1986 JANNAF Prqjulsion Meeting, New Orleans, Louisiana, 26-28 August 1986. 120 W. W. Smith, S. C. Knowles, “Analysis of Electric Propulsion Concepts for Near- term Mission i^jplication,” 1985 JANNAF 168 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Propulsion Meeting , San Diego, California 9-12 April 1985. 121 J. J. Bless, A. M. Sutton, “Integration And Verification Of A 30 kW Arcjet Spacecraft System,” Paper AIAA-94-3143, 30* Joint Propulsion Conference, Indianapolis, IN, 27- 29 Jun 1994. 122 J. A Pobst, J. H. Schilling, F. M. Lufty, D. A Erwin, R. Spores, , “Arcjet Diagnostics Using Current Modulation Velocimetry And Pulsed Electron Beam Fluorescence," Paper AIAA-94-2742, , 30* Joint Propulsion Conference, Indianapolis, IN, 27-29 Jun 1994. 1 2 3 D. Perkins, “The Air Force Electric Propulsion Program,” Paper AIAA-94-2734, 30* Joint Propulsion Conference, Indianapolis, IN, 27-29 Jun 1994. 1 2 4 J. Schilling, J. Pobst, D. Erwin, “The Use of Pulsed Electron Beam Fluorescence for Arcjet Plume Diagnostics,” Paper IEPC-93- 130, 23"^ International Electric Propulsion Conference, Seattle, WA, 13-18 Sept 1993. 1 2 5 J. Pobst, J. Schilling, D. Erwin, R. Spores, “Time Resolved Measurements of IkW Arcjet Plumes Using Current Modulation Velocimetry and Triple Langmuir Probes,” Paper IEPC-93-128, 23* International Electric Propulsion Conference, Seattle, WA, 13-18 Sept 1993. 126 A M. Sutton, “Overview of the Air Force ESEX Flight Experiment,” Paper IEPC-93- 057, 23* International Electric Propulsion Conference, Seattle, WA, 13-18 Sept 1993. 1 2 7 J. M. Jones, R. S. Einhom, “The USAF Electric Propulsion Systems Activities, 1993,” Paper AIAA-93-1937,29* Joint Propulsion Conference, Monterey, CA, 28- 30 Jun 1993. 128 D. L. Tilley, K. A McFall, S. Castillo, J. C. Andrews, “An Investigation Of The Breakdown Voltage Characteristics Of A 30 kW Class Ammonia Arcjet,” Paper AlAA- 93-1901, 29* Joint Propulsion Conference, Monterey, CA, 28-30 June 1993. 1 2 9 D. R. Bromaghim, “Measurements Of Magnetic Fields Produced By A 30 kWe Class Arcjet Power Conditioning Unit,” Final Report PL-TR-92-3049, Jan 1993. 1 3 0 J. M. Jones, T. M. Sanks, B. L. Raygor, “The USAF Electric Propulsion Systems Activities,” Paper AlAA-92-3702, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 13 1 R. A Spores, J. A Pobst, J. H. Schilling, D. A Erwin, “Performance Effects Of Interaction Between A Low-Power Arcjet And Its Power Processing Unit,” Paper AIAA-92-3238, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 1 3 2 K. D. Hicks, “An Overview Of Electric Propulsion Technology For Earth-Orbiting Missions,” Final Report Jan-Jun 92, PL-TR- 92-3037 XPX-92002. 1 3 3 R. Spores, D. Erwin, “Investigation Of Arcjet-Power Processing Unit Interaction,” Contractor’s Meeting in Propulsion, Air Force Office of Scientific Research, San Diego, CA, 15-19 June 1992. 1 3 4 D. Bromaghim, “Orbital Transfer Systems Analysis,” Final Report 11 May 88-15 Apr 91. PL-TR-91-3087 XPX-92002. 1 3 5 S. Castillo, “Arcjet Facility,” Final Report,. Aug 86-Aug 91 PL-TR-91-3085. 1 3 6 W. M. Schmidt, “Analysis Of Expendable Solar Electric Orbit Transfer Vehicles,” CPIA-PUB-550-VOL-2, 1990 JANNAF Propulsion Meeting, Anaheim, CA, 3-5 Oct 1990 169 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1 3 7 K. Ennix, W. Pritz, "A System Level Analysis Of The Ammonia Arcjet," CPIA- PUB-550-VOL-2, 1990 JANNAF Propulsion Meeting, Anaheim, CA, 3-5 Oct 1990. 1 3 8 J. S. Maher, J. C. Andrews, “The Analysis Of Arcjet Cathodes By Scanning Electron Microscopy,” Paper AIAA-91- 2452, 27* Joint Propulsion Conference, Sacramento, CA, 24-26 Jim 1991. 1 3 9 R. Meya, T. Nguyen, “ 1987 Status Report: United States Air Force Electric Propulsion Research And Development,” Paper AIAA-87-1036,19* International Electric Propulsion Conference, Colorado Springs, CO, 11-13 May 1987. 1 4 0 F. B. Mead, “Future Possibilities For Nonconventional Propulsion Developments At The Air Force Rocket Propulsion Laboratory,” Vol-1, 1986 JANNAF Propulsion Meeting, New Orleans, Louisiana, 26-28 August 1986. 1 4 1 K. D. Goodfellow, J. E. Polk, “Design and Development of a 3- to 10- kW Ammonia Arcjet,” Paper ŒPC-93-078, 23"^ International Electric Propulsion Conference, Seattle, WA, 13-17 Sept 1993. 1 4 : J. E. Polk, K. D. Goodfellow, “Results Of A 1462 Hour Ammonia Arcjet Endurance Test,” Paper AIAA-92-3833, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 1 4 3 J. E. Polk, K. D. Goodfellow, “Endurance Test of an Ammonia Arcjet at lOkWe," Paper Œ PC-91-068, 22“* International Electric Propulsion Conference, Viareggio, Italy, 14-18 October 1991. 1 4 4 W. D. Deininger, A. Chopra, T. J. Pivirotto, K. D. Goodfellow, J. W. Barnett,, “30-Kw Ammonia Arcjet Technology,” Fitial Report, Jul 86-Dec 89, JPL-Publ-90-4 NASA-CR-186850, Jan 1990. 1 4 5 B. Palaszewski, “Electric Propulsion For Manned Mars Exploration,” CPIA-PUB- 5 15-VOL-l, 1989 JANNAF Propulsion Meeting, Cleveland, OH, 23-25 May 1989. 1 4 6 C. Gamer, “Mission Options Few An Electric Prc^ulsion Demonstration Flight Test,” CPIA-PUB-515-VOL-l, 1989 JANNAF Propulsion Meeting, Cleveland, OH, 23-25 May 1989. 1 4 7 W. D. Deininger, R. J. Vondra, W. J. Schafer, “Spacecraft And Mission Design For The Space Nuclear Power Source Reference Mission,” Paper AIAA-87-2026, 23"* Joint Propulsion Conference, San Diego, CA, 29 Jun-2 Jul, 1987. 1 4 8 B. Palaszewski, C. Engelbrecht, “Lightweight Spacecraft Propulsion System Selection,” Paper AIAA-87-2022, 23"' Joint Propulsion Conference, San Diego, CA, 29 Jun-2 Jul, 1987. 1 4 9 T. J. Pivirotto, D. Q. King, W. D. Deininger, “Long Duration Test Of A 30-Kw Class Thermal Arcjet Engine,” Paper AIAA- 87-1947, 23"* Joint Propulsion Conference, San Diego, CA, 29 Jun-2 Jul, 1987. 1 5 0 T. J. Pivirotto, W. D. Deininger, “Velocity Measurements In The Plume Of An Arcjet Engine,” Paper AIAA-87-1063, 19* International Electric Propulsion Conference, Colorado Springs, CO, 11-13 May 1987. 1 5 1 W. D. Deininger, R. J. Vondra, “Design And Performance Of An Arcjet Nuclear Electric Propulsion System For Amid-1990s Reference Mission,” Paper AIAA-87-1037, 19* International Electric Propulsion Conference, Colorado Springs, CO, 11-13 May 1987. 170 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1 5 2 B. Palaszewski, “Geosynchronous Earth Orbit Base Propulsion: Electric Propulsion Options,” Paper AIAA-87-0990, 19* International Electric Propulsion Conference, ColOTado Springs, CO, 11*13 May 1987. 1 5 3 T. J. Pivirotto, D. Q. King, J. R. Brophy, W. D. Deininger, “Performance And Long Duration Test Of A 30-Kw Thermal Arcjet Engine,” Final Report, 28 Dec 84-30 Nov 86 JPL-D^643 AFAL-TR-87-010, Dec 1986. 1 5 4 B. A. Palaszewski, “Resistojet- And Arcjet-Propulsion Power Processing,” Vol. 1, 1986 JANNAF Propulsion Meeting, New Orleans, Louisiana, 26-28 August 1986. 1 5 5 B. A. Palaszewski, “Advanced Propulsion For Polar-Orbiting And Co-Orbiting Free Flyers,” Paper AlAA-86-1564, 22“* Joint Propulsion Conference, Huntsville, AL, 16- 18 June 1986. 1 5 6 W. Deininger, R. Vondra, “Development Of An Arcjet Nuclear Electric Propulsion System For A 1993 Flight,” Paper AIAA-86- 1510, 22“* Joint Propulsion Conference, Huntsville, AL, 16-18 June 1986. 1 5 7 T. Pivirotto, D. Q. King, W. Deininger, J. R. Brophy, “The Design And Operating Characteristics Of A 30 kW Thermal Arcjet Engine For Space Propulsion,” Paper AlAA- 86-1508, 22“* Joint Propulsion Conference, Huntsville, AL, 16-18 June 1986. 1 5 8 T. J. Pivirotto, D. Q. King, “Thermal Arcjet Technology for Space Propulsion,” Vol. 1, 1985 JANNAF Propulsion Meeting , San Diego, CA, 9-12 April 1985. 1 5 9 T. J. Pivirotto, D. Q. King, J. R. Brophy, “Development And Life-Testing Of 10 kW Class Thermal Arcjet Engines,” Paper AlAA-85-2031, 18* International Electric Propulsion Conference, Alexandria, VA, 30 Sep - 2 Oct 1985. 160 J. R. Brophy, T. J. Pivirotto, D. Q. King, “Investigation Of Arcjet Nozzle Performance,” Paper AlAA-85-2016, 18* International Electric Propulsion Conference, Alexandria, VA, 30 Sep - 2 Oct 1985. 161R. M. Jones, “ A Comparison Of Potential Electric Propulsion Systems For Orbit Transfer,” Paper AlAA-82-1871,16* International Electric Propulsion Conference, New Orleans, LA, 17-19 November 1982. 162 M. Riehle, M. Auweter-Kurtz, H. L. Kurtz, “Perfcrmance Investigations Of Radiation And Regeneratively Cooled 5 kW Thermal Arcjet Thruster,” Paper AlAA-94- 3123, 30* Joint Propulsion Conference, Indianapolis, IN, 27-29 Jun 1994. 1 6 3 D. M. Zube, E. W. Messerschmid, “Spectrosccpic Temperature And Density Measurements In A Low Power Arcjet Pltime,” Paper AIAA-94-2744, 30* Joint Propulsion Conference, Indianapolis, IN, 27- 29 Jun 1994. i«E. W. Messerschmi, D. M. Zube, H. L. Kurtz, K. Meinzer, “Development and Utilization Objectives of a Low-Power Arcjet for the P3D (OSCAR) Satellite,” Paper lEPC-93-056, 23”* International Electric Propulsion Conference, Seattle, WA, 13-17 Sept 1993. 1 6 5 D. M. Zube, M. Auweter-Kurtz, “Spectroscopic Arcjet Diagnostic Under Thermal Equilibrium And Nonequilibrium Conditions,” Paper AlAA-93-1792, 29* Joint Propulsion Conference, Monterey, CA, 28-30 June 1993. 166 D. M. Zube, E. W. Messerschmid, “ATOS Flight Experiment For A 700 W Ammonia Arcjet,” Paper AIAA-93-2224, 29* Joint Prcpulsion Conference, Monterey, CA, 28-30 Jun 1993. 171 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1 6 7 M. Riehle, B. Glocker, M. Auweter- Kurtz, H. Kurtz, •‘Development O f A High Specific 1.5 To 5 kW Thermal Arcjet,” Report, IRS-93-P7 NASA-CR-193I33, 1993. 168 T. M. Golz, M. Auweter-Kurtz, H. L. Kurtz, “ 100 kW Hydrogen Arcjet Thruster Experiments," Paper AIAA-92-3836, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 1 6 9 B. Glocker, M. Auweter-Kurtz, “Numerical And Experimental Constrictor Flow Analysis Of A 10 kW Thermal Arcjet,” Paper AIAA-92-3835, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 1 7 0 B. Glocker, M. Auweter-Kurtz, “Radiation Cooled Medium Power Arcjet Experiments And Thermal Analysis,” Paper AIAA-92-3834, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 1 7 1 H. L. Kurtz, D. M. Zube, B. Glocker, M. Auweter-Kurtz, M. Kiimersley, “Low Power Hydrazine Arcjet Thruster Study,” Paper AIAA-92-3116, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 1 7 2 E. Tosti, H. O. Schrade, C. Petagna, “Comparison Of 15-Kwe Water-Cooled Arcjet Test Results At Two Different Facilities,” AIAA Journal of Propulsion and Power, ^ 4, 911-914, Jul-Aug 1992. 1 7 3 T. M. Goelz, M. Auweter-Kurtz, H. L. Kurtz, H. O. Schrade, “High Power Arcjet,” Report, Feb 9 1-Feb 92, NASA-CR-190238, March 1992. 1 7 4 D. M. Zube, B. Glocker, H. L. Kurtz, E. W. Messerschmid, “Arcjet Test Missions (ARTEMIS) on EURECA,” Paper ŒPC-91- 155, 22“* International Electric Propulsion Conference, Viareggio, Italy, 14-18 October 1991. 1 7 5 Th. Golz, M. Auweter-Kurtz, H. L. Kurtz, “High Power Arcjet Thruster Experiments,” Paper IEPC-91-072, 22“* International Electric Propulsion Conference, Viareggio, Italy, 14-18 October 1991. 1 7 6 D. M. Zube, B. Glocker, H. L. Kurtz, M. Kiimersley, G. Matthaus, “Development of a Low Power Radiatively Cooled Thermal Arcjet Thruster,” Paper ŒPC-91-042, 22“* International Electric Propulsion Conference, Viareggio, Italy, 14-18 October 1991. 1 7 7 B. Glocker, Th. Rosgen, A Laxander “Medium Power Acjet Analysis and Experiments,” Paper ŒPC-91-016, 22“ “ * International Electric Propulsion Conference, Viareggio, Italy, 14-18 October 1991. 1 7 8 D. M. Zube, R. M. Myers, “Nonequilibrium In A Low Power A cjet Nozzle,” Paper AIAA-91-2113, 27* Joint Propulsion Conference, Sacramento, C A 24- 26 Jun 1991.’ 1 7 9 T. M. Goelz, M. Auweter-Kurtz, H. L. Kurtz, H. O. Schrade, “High Power A cjet,” Report, Aug 90-Jan 91 NASA-CR-187986, Feb 1991. 180 J. E. Pollard, “Acjet Plume Studies Using Molecular Beam Mass Spectrometry,” Paper IEPC-93-132, 23"* International Electric Propulsion Conference, Seattle, WA 13-17 Sept 1993. 181L. K. Johnson, A Rivera, M. Lundquist, T. M. Sanks, A Sutton, D. R. Bromaghim, “Frequency-Domain Electromagnetic Characteristics Of A 26 kW Ammonia A cjet,” Paper AIAA-93-2393, 29* Joint Propulsion Conference, Monterey, C A 28- 30 June 1993. 182 J. E. Pollard, D. E. Jackson, D. C. Marvin, A. B. Jenkin, S. Janson, “Electric Propulsion Flight Experience And Technology Readiness,” Paper AIAA-93- 172 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2221, 29* Joint F*ropuision Conference, Monterey, CA, 28-30 Jun 1993. 1 8 3 M. W. Crofton, “Advanced Diagnostic Techniques For Electric Propulsion,” Paper AIAA-93-1794,29* Joint Propulsion Conference, Monterey, CA, 28-30 Jun 1993. i8iS. W. Janson, “The Impact Of Advanced Diagnostic Techniques On Electrothermal Thruster Design,” Paper AIAA-92-3242, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 1 8 5 M. W. Crofton, R. P. Welle, S. W. Janson, R. B. Cohen, “Temperature, Velocity And Density Studies In The 1 kW Ammonia Arcjet Plume By LIF,” Paper AIAA-92- 3241,28* Joint Propulsion Conference, NashvUle, TN, 6-8 Jul 1992. 186 M. W. Crofton, “Spectral Irradiance Of The 1 kW Arcjet Thruster From 80 To 500 Nm,” Paper AIAA-92-3237, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 1 8 7 E. J. Belting, “Two Dimensional Coherent Anti-Stokes Raman Scattering With Application To The Hydrogen Arcjet,” Contractor’s Meeting in Propulsion, Air Force Office of Scientific Research, San Diego, CA, 15-19 June 1992. 188 J. W. Ramos, D. M. Sutton, M. G. McLain, K. P. Zondervan, “Electric Propulsion Applications,” CPIA-PUB-580- VOL-1, 1992 JANNAF Propulsion Meeting, Indianapolis, IN, 24-27 Feb 1992. 1 8 9 M. W. Crofton, R. P. WeUe, S. W. Janson, R. B. Cohen, “Advanced Diagnostics of Arcjets,” Paper IEPC-91-095, 22“* International Electric Propulsion Conference, Viareggio, Italy. 14-18 October 1991. 1 9 0 R. P. Welle, J. E. Pollard, S. W. Janson, M. W. Crofton, R. Cohen, , “One Kilowatt Hydrogen And Helium Arcjet Performance,” Paper AIAA-91 -2229,27"" Joint FTopulsion Conference, Sacramento, CA, 24-26 Jun 1991. 1 9 1 J. E. Pollard, S. W. Janson, D. A. Lichtin, R. B. Cohen, “Ion Beam Characterization Using The MVA, And Capabilities Of The Aerospace Advanced Propulsion Diagnostic Facility,” CPIA-PUB-515-VOL-l, 1989 JANNAF Propulsion Meeting, Cleveland, OH, 23-25 May 1989. 1 9 : D. A. Lichtin, J. E. PoUard, S. W. Janson, R. B. Cohen, “A Mass/Velocity Analyzer For Electric Thruster Plumes: Initial Results,” CPIA-PUB-+80-VOL-1, 1987 JANNAF Propulsion Meeting, San Diego, CA, 15-17 Dec 1987. 1 9 3 J. E. Pollard, D. A. Lichtin, R. B. Cohen, “RF (Radio Frequency) Discharge Electrothermal Propulsion: Results From A Lab-Scale Thruster,” Paper AIAA-87- 2124,23"' Joint Propulsion Conference, San Diego, CA, 29 Jun-2 Jul, 1987. m J. M. Sponable, J. P. Penn, “Electric Propulsion For Orbit Transfer - A NAVSTAR Case Study,” Paper AIAA-87- 0985, 19* International Electric Propulsion Conference, Colorado Springs, CO, 11-13 May 1987. 1 9 5 D. A. Lichtin, R. B. Cohen, J. E. Pollard, R. Rianda, “Advanced Diagnostic Thruster Evaluation Facility FY 85 Annual Report,” Annual Report, Oct 84-Oct 85. 1 9 6 W. D. Deininger, R. Di Stefano, G. Parisi, E. Detoma, G. Botto, “A Review Of Low Power Arcjet PCU Development And Testing,” Paper AIAA-94-3008,30* Joint Propulsion Conference, Indianapolis, IN, 27- 29 Jun 1994. 1 9 7 E. Tosti, W. D. Deininger, “Emission Spectroscopy of IkWe Arcjet Operating with 173 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Simulated Hydrazine," Paper IEPC-93-I34, 23"^ International Electric Propulsion Conference, Seattle, WA 13-17 Sept 1993. 1 9 8 W. D. Deininger, M. Andrenucci, G. Saccoccia, “A Review of the ESA ASTP-3 MPD/ Arcjet Development Program (1988- 1993),” Paper IEPC-93-081, 23"^ Intemaüonal Electric Propulsion Conference, Seattle, WA 13-17 Sept 1993. 1 9 9 W. D. Deininger, M. Vulpiani, E. Tosti, R. Di Stefano, E. Detoma, S. Ferrari, M. Rossi, “Low Power Arcjet System Description for North-South Station Keeping,” Paper IEPC-93-012, 23"" International Electric Propulsion Conference, Seatüe, WA 13-17 Sept 1993. 200 W. D. Deininger, M. Attili, M. Vulpiani, E. Detoma, “Low Power Arcjet System Requirements Definition For North-South Stationkeeping,” Paper AIAA-93-2223, 29^ Joint Propulsion Conference, Monterey, CA,28-30 Jun 1993. 201W. D. Deininger, M. Andrenucci, “A Review Of Electric Propulsion Activities At BPD Difesa E Spazio And Centrospazio,” Paper AIAA-93-1938, 29* Joint Propulsion Conference, Monterey, CA, 28-30 June 1993. 202 G. Cruciani, W. D. Deininger, “A 300- Hour Endurance Test Of An Arcjet Thruster At 1 kW,” Paper AIAA-93-1900, 29* Joint Propulsion Conference, Monterey, CA, 28- 30 June 1993. 2 0 3 W. D. Deininger, M. J. Glogowski, G. Cruciani, F.Scortecci, “Comparison Of Low Power Arcjet Operation At BPD And Centrospazio,” Paper AIAA-92-3115, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 2 0 4 G. Cruciani, W. D. Deininger, “Development Testing Of A IkW Class Arcjet Thruster,” Paper AIAA-92-3114, 28* Joint Propulsion Conference, Nashville, TN. 6-8 Jul 1992. 2 0 5 E. Tosti, W. D. Deininger, “Plume Analysis of A lOkWe/N Arcjet Using Emission Spectroscopy,” Paper IEPC-91-092, 22'"' International Electric Propulsion Conference, Viareggio, Italy, 14- 18 Oct 1991. 206 M. Andrenucci, G. Saccoccia, F. Scortecci, N. Panattoni, U. Schulz, W. D. Deininger, “Performance Study of a Laboratory Model of Low Power Arcjet,” Paper IEPC-91-045, 22'"' International Electric Propulsion Conference, Viareggio, Italy, 14-18 Oct 1991. 2 0 7 G. Cruciani, M.J. Glogowski, W.D. Deininger, A. Trippi, “Advanced Laboratory Model, 1 kW-Class Arcjet Engine Testing,” Paper IEPC-91-044, 22'* International Electric Propulsion Conference, Viareggio, Italy, 14-18 Oct 1991. 208 E. Tosti, B. Glocker, A. Trippi, “Final Results of 15kWe Water Cooled Arcjet Testing,” Paper IEPC-91-014, 22'* International Electric Propulsion Conference, Viareggio, Italy, 14-18 Oct 1991. 2 0 9 W. D. Deininger, “Review Of Arcjet Technology Development At BPD,” Paper AIAA-91-1992, 27* Joint Propulsion Conference, Sacramento, CA, 24-26 Jun 1991. 210 P.V. Storm and M .A Cappelli, “High Spectral Resolution Emission Study of a Low Power Hydrogen Arcjet Plume,” Paper AIAA 95-1960, 26* Plasmadynamics and Lasers Conference, San Diego, California, 19-22 June, 1995. 22 211 J. G. Liebeskind, R. K. Hanson, M. A Cappelli, “Experimental Investigation Of 174 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Velocity Slip Near An Arcjet Exit Plane.” AIAA Journal, 33, 373-5. Feb 1995. 2i:M. H. Loh, J. G. Liebeskind, M. A Cappelli, “Arcjet Thrusters For The Synthesis Of Diamond And Cubic-Boron Nitride FUms,” Paper AIAA-94-3233,30* Joint Propulsion Conference, Indianapolis, IN, 27-29 Jun 1994. :nP. V. Storm, M. A. Cappelli, “Axial Emission Measurements On A Medium Power Hydrogen Arcjet Thruster,” Paper AIAA-94-2743,39* Joint Propulsion Conference, Indianapolis, IN, 27-29 Jun 1994. :wP. Storm, M. Cappelli, “Axial Emission Diagnostics of a Low Power Hydrogen Arcjet Thruster," Paper IEPC-93-219, 23"* International Electric Propulsion Conference, Seattle, WA, 13-17 Sept 1993. 2 1 5 M. Cappelli, M. Loh, J. Liebeskind, “Use of Hydrogen Arcjet Thrusters for Diamond Synthesis,” Paper IEPC-93-166, 23"* International Electric Propulsion Conference, Seattle, WA, 13-17 Sept 1993. 216 J. G. Liebeskind, R. K. Hanson, M. A. Cappelli, “Plume Characteristics Of An Arcjet Thruster," Paper AIAA-93-2530, 29* Joint Propulsion Conference, Monterey, CA, 28-30 June 1993. 2 1 7 M. H. Loh, J. G. Liebeskind, M. A Cappelli, “Characterization Of A Supersonic Hydrogen Arcjet Plasma Thruster Employed In Diamond Film Synthesis,” Paper AIAA- 93-2227, 29* Joint Propulsion Conference, Monterey, CA, 28-30 Jun 1993. 218 M. H. Loh, M. A Cappelli, “Supersonic DC-Arcjet Plasma At Sub torr Pressures As A Medium For Diamond Filmsynthesis,” Paper AIAA-92-3534, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 2 1 9 J. G. Liebeskind, R. K. Hanson, M. A Cappelli, “Flow Diagnostics Of An Arcjet Using Laser-Induced Fluorescence,” Paper AIAA-92-3243, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 220 M. A Cappelli, “Modeling Of The Near- Electrode Regions Of Arcjets I: Coupling Of The Flowfield to The Non-Equilibrium Boundary Layer," Paper AIAA-92-3109, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 2 21M. A Cappelli, R. K. Hanson, J. G. Liebeskind, D. H. Manzella. “Optical Diagnostics of a Low Power Hydrogen Arcjet,” Paper IEPC-91-091, 22“ * International Electric Propulsion Conference, Viareggio, Italy, 14-18 Oct 1991. 222 J. G. Liebeskind, R. K. Hanson, M. A Cappelli, “Velocity Measurements In A Hydrogen Arcjet Using LIF,” Paper AIAA- 91-2112, 27* Joint Propulsion Conference, Sacramento, CA, 24-26 Jun 1991. 2 2 3 D. Burtner, D. Keefer, W. Ruyten, “Exf>erimental And Numerical Studies Of A Low-Power Arcjet Operated On Simulated Ammonia,” Paper AIAA-94-2869, 30* Joint Propulsion Conference, Indianapolis, IN, 27- 29 Jun 1994. 2 2 4 Ruyten, W. M. Keefer, D. “Two-Beam Multiplexed Laser-Induced Fluorescence Measurements Of An Argon Arcjet Plume,” AIAA Journal, 3 1 2083-9,Nov 1993. 2 2 5 W. M. Ruyten, D. Burtner, D. Keefer, “Spectroscopic Investigation Of A Low- Power Arcjet Plume,” Paper AIAA-93-1790, 29* Joint Propulsion Conference, Monterey, CA, 28-30 June 1993. 226 D. Keefer, R. Rhodes, “Arcjet Space Thrusters," Final Technical Report, Mar 90- May 93 NASA-CR-193191. 175 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. =7 V. J. Friecfly, G. W. Garrison, W. M. Ruyten, “EPOP: Toward The Realization Of An Electric Propulsion Orbital Platform,” Paper AIAA-92-3201, 28“* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. nsT. Moeller, R. Rhodes, D. Keefer, A. Sedghi-Nasab, W. M. Ruyten, “Comparison Of Experimental And Numerical Results Po t An Argon Arcjet,” Paper AIAA-92-3105, 28“’ Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. =9 Keefer, D. "LIE Diagnostics For Arcjet Thrusters,” Contractor’s Meeting in Propulsion, Air Force Office of Scientific Research, San Diego, CA, 15-19 Jun 1992. boR. Rhodes, D. Keefer, “Comparison of Model Calculations With Experimental Data From Hydrogen Arcjets,” Paper ŒPC-91- 111, 22“ “ International Electric Propulsion Conference, Viareggio, Italy, 14-18 Oct 1991. Bi W. M. Ruyten, D. Keefer, “Laser Fluorescence Velocimetry of an Arcjet Exhaust Plume,” Paper ŒPC-91-093. 22” “ International Electric Propulsion Conference, Viareggio, Italy, 14-18 Oct 1991. B:R. Rhodes, D. Keefer, “Modeling Arcjet Space Thrusters,” Paper AIAA-91-1994, 27“’ Joint Propulsion Conference, Sacramento, CA, 24-26 Jun 1991. 2 3 3 M. Martinez-Sanchez, “Performance Potential Of Plasma Thrusters: Arcjet And Hall Thruster Modeling,” Final Report. 1 May 92 - 30 Apr 93 AFOSR-TR-93-0753. 2 3 4 S. A. Miller, M. Martinez-Sanchez, “Multifold Nonequilibrium Simulation Of Electrothermal Arcjets,” Paper AIAA-93- 2101, 29* Joint Propulsion Conference, Monterey, CA, 28-30 June 1993. 2 3 5 M. Martinez-Sanchez, A. Sakamoto, “Simplified Analysis Of Arcjet Thrusters,” Paper AIAA-93-1904, 29* Joint Propulsion Conference, Monterey, CA, 28-30 June 1993. 2 3 6 M. Martinez-Sanchez, “Performance Characteristics Of Plasma Thrusters,” Contractor’s Meeting in Propulsion, Air Force Office of Scientific Research, San Diego, CA, 15-19 June 1992. 2 3 7 S. Miller, M. Martinez-Sanchez, “Viscous and Diffusive Effects in Electrothermal and MPD Arcjet Thrusters,” Paper ŒPC-91- 060, 22“ “ International Electric Propulsion Conference, Viareggio, Italy, 14-18 Oct 1991. 2 3 8 D. J. Heimerdinger, M. Martinez-Sanchez, J. F. Davis, P. J. Turchi, “Effect Of Axial Variation Of Electrode Spacing On MPD (Magnetoplasmadynamic) Arcjet Behavior,” Paper AIAA-87-0998, 19* International Electric Propulsion Conference, Colorado Springs, CO, 11-13 May 1987. 2 3 9 M. Martinez-Sanchez, D. Heimerdinger, ‘Two-Dimensional Analysis Of An Arcjet,” Paper AIAA-85-2040, 18* International Electric Propulsion Conference, Alexandria, VA, 30 Sep - 2 Oct 1985. 2 4 0 W. J. Harris, E. A. O’Hair, L. L. Hatfield, M. Kristiansen, “Cathode Erosion on Medium to High Power Arcjet Thrusters,” Paper ŒPC-93-028, 23’ “ International Electric Propulsion Conference, Seattle, WA, 13-17 Sept 1993. 2 4 1 J. s. Mankins, W. J. Harris, E. A. O’Hair, L. L. Hatfield, M. Kristiansen, “Comparison Of Erosion Of Various Cathode Materials In A 30 kWe Class Arcjet, ” Paper AIAA-92- 3839, 28* Joint Propulsion Conference, NashvUle, TN, 6-8 Jul 1992. 176 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2j:W . j. Harris, E. A. O'Hair, L. L. Hatfield, M. Kristiansen, J. Mankins, "Anode Arc Motion In High Power Arcjets,” Paper AIAA-92-3838, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. W 3 W. J. Harris, E. A. O’Hair, L. L. Hatfield, M. Kristiansen, “Physical Model And Experimental Results Of Cathode Erosion Related To Power Supply Ripple," Paper AIAA-92-3837, 28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. luM . D. Grimes, W. J. Harris, E. A. O ’Hair, L. L. Hatfield, M. Kritiansen, M. C. Baker, “Acoustical Resonant Behavior and V-I Curves for a 30kWe Nitrogen Arcjet,” Paper IEPC-91-071, 22°“ ' International Electric Fh-opulsion Conference, Viareggio, Italy, 14-18 Oct 1991. 2 4 5 M. D. Grimes, W. J. Harris, E. A. O ’Hair, L. L. Hatfield, M. Kristiansen, M. Baker, “Continuous Characteristic V-I Curves For A 30 kWe Nitrogen Arcjet,” Paper AIAA-91-2225 , 27* Joint Propulsion Conference, Sacramento, CA 24-26 Jun 1991. 2 4 6 R. L. Burton, S. A. Button, N. T. Tiliakos, H. Krier, “Application O f Multiple Electrostatic Probes To A Low Power Arcjet,” Paper AIAA-94-3299,30* Joint Propulsion Conference, Indianapolis, IN, 27- 29 Jun 1994. 2 4 7 G. F. Willmes, R. L. Burton, “Thrust Performance Of A Very Low Power Pulsed Arcjet,” Paper AIAA-94-3125, 30* Joint Propulsion Conference, Indianapolis, IN, 27- 29 Jun 1994. 2 4 8 G. Willmes, R. Burton, “Investigation of a Very Low Power Pulsed Arcjet,” Paper ŒPC-93-136, 23"* International Electric Propulsion Conference, Seattle, WA, 13-17 Sept 1993. 2 4 9 R. D. Taylor, R. L. Burton, K. K. Wetzel, “Preliminary Investigation Of A Low Power Pulsed Arcjet Thruster,” Paper AIAA-92- 3113,28* Joint Propulsion Conference, Nashville, TN, 6-8 Jul 1992. 2 5 0 V. Babu, S. Aithal, V. V. Subramaniam, “On the Effects of Swirl in Arcjet Thruster Flows,” Paper ŒPC-93-183, 23* International Electric Propulsion Conference, Seattle, WA, 13-17 Sept 1993. 25 1 V. Babu, S. Aithal, V. V. Subramaniam, “Vibrational Non-Equilibrium in Arcjet Flows,” Paper IEPC-93-129, 23* International Electric Propulsion Conference, Seattle, WA, 13-17 Sept 1993. 2 5 2 V. Babu, S. Aithal, V. V. Subramaniam, “Recovery Of Frozen Row Losses In Arcjets,” Final Report, 1 Jul 91-30 Sep 93 RF725071-769221 AFOSR-TR-94-0259. 2 5 3 V. V. Subramaniam, “Recovery Of Frozen Row Losses In Arcjets,” Contractor’s Meeting in Propulsion, Air Force Office of Scientific Research, San Diego, CA, 15-19 June 1992. 2 5 4 S. Ohler, B. Gilchrist, A D. GaUimore, “Non-Intrusive Plasma Diagnostics For Electric propulsion Research, ” Paper AIAA- 94-3297, 30* Joint Propulsion Conference, Indianapolis, IN, 27-29 Jun 1994. 2 5 5 A D. GaUimore, S.-W. Kim, J. E. Foster, L. B. King, F. S. Gulczinski, “Near And Far- Field Plume Studies Of A 1 kW Arcjet,” Paper AIAA-94-3137, 30* Joint Propulsion Conference, Indianapolis, IN, 27-29 Jun 1994. 2 5 6 J. E. M. Goldsmith, J. A MiUer, R. J. M. Anderson, and L. R. Williams, “Multiphoton-Excited Ruorescence Measurements Of Absolute Concentration profiles Of Atomic Hydrogen In Low- iressm’ e Rames,” Proceedings of the 23* 177 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Symposium (International) on Combustion, Combustion Institute, 1821, 1990. ■ 'jsil. E. M. Goldsmith and L. A. Rahn, “Doppler-Free Two-Photon-Excited Fluorescence Spectroscopy of Atomic Hydrogen in Flames,” Optics Letters 14, 814-816, 1990. 2 5 8 J. E. M. Goldsmith, “Investigation of Simultaneous Two-Photon H and Single Photon OH Excitation In Flames Using a Single Dye Laser,” Applied Optics, 29, 33, 4841-4842,1990. 2 5 9 J. E. M. Goldsmith, R. J. M. Anderson, and L. R. Williams, “Time-Resolved Two- Photon-Excited Fluorescence Detection of Atomic Hydrogen in Flames,” Optics Letters 15, 1,78-80, 1990. 260 U. Westblom, S. Agrup, M. Alden, H. M. Hertz, and J. E. M. Goldsmith, “Properties of Laser-Induced Stimulated Emission for Diagnostic Purposes,” Applied Physics B 50, 487-497, 1990. 261 J. E. M. Goldsmith, “Two-Step Saturated Fluorescence Detection Of Atomic Hydrogen In Flames,” Opt. Leners 10. 116. 1985. 262 U. Czametzki, K. Miyazaki, T. Kajiwara, and K. Miuaoka, M. Maeda, and H. F. Dobele, “Comparison of Various Two- Photon Excitation Schemes for Laser- Induced Fluorescence Spectroscopy in Atomic Hydrogen,” J. Opt. Soc. Am. B. Ü , 11,2155-2162, 1994. 2 6 3 K. Muraoka and M. Maeda, “Application of Laser-Induced Fluorescence to High- Temperature Plasmas,” Plasma Phys. Conuol. Fusion 35, 633-656, 1993. 264?. V. Storm and M. A. Cappelli, “Stark Broadening Corrections to Laser-Induced Fluorescence Temperature Measurements in a Hydrogen Arcjet Plume,” J. Applied Optics 35, 24,4913-4918, 1996. 2 6 5 K.P. Huber and G. Herzberg, Molecular Spectra and Molecular Structure: IV. Constants o f Diatomic Molecules, Van Nostrand Reinhold Co., NY, 1979. 178 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. IMAGE EVALUATION TEST TARGET (Q A -3) / y f % 1 . 0 i.i 1 .2 5 !M 2.2 1.4 2.0 1 . 8 1 . 6 150m m 6 " V c P . /, o / / / yIP P L IE D ^ IIVWGE . Inc 1653 East Main Street Rochester, NY 14609 USA Phone: 716/482-0300 — — Fax; 716/288-5989 O 1993. Applied Image. Inc.. Ail Rights Reserved R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Energy distribution in the plume of a helium arcjet

PDF

Large eddy simulation of turbulent convection

PDF

Plasma species prediction using direct simulation Monte Carlo (DSMC)

PDF

A theoretical and experimental investigation of cathode processes in electric thrusters

PDF

Vortex motion and the geometric phase

PDF

Thermodynamic and kinetic properties of some diatomic gases

PDF

Instability and flow vectoring of jets

PDF

Convective instability of fluid interfaces

PDF

Laser wakefield accelerators and laser instabilities in a hollow plasma channel

PDF

Multiphoton ionization of the hydrogen atom and the hydrogen molecular ion

PDF

Experimental investigation of boundary layer ingestion into diffusing inlets.

PDF

Instabilities of propagating quasi-2D gaseous flames and chemical fronts in narrow channels.

PDF

Experimental studies of the interactive /dynamics of Hele -Shaw flows and directional solidification

PDF

Dynamics of heavy particles within strained vortical flow structures

PDF

Strong helioseismic constraints on weakly-coupled plasmas

PDF

Static and dynamic leading-edge-flap effects on the vortex lift of a delta wing.

PDF

Theoretical, computational, and experimental investigation of tunable radiation sources based on relativistic ionization fronts

PDF

Fluorophore characterization using spectro-temporal laser-induced fluorescence methods

PDF

Scattering of light by turbulence

PDF

Multi-cavity lasers

Asset Metadata

Creator Pobst, Jeffrey Alan (author)

Core Title Multiphoton laser-induced fluorescence for measuring point specific densities of ground state atomic hydrogen in an arcjet plume

Degree Doctor of Philosophy

Degree Program Aerospace Engineering

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

Tag engineering, aerospace,OAI-PMH Harvest,Physics, Fluid and Plasma,physics, optics

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-331103

Unique identifier UC11353843

Identifier 9835117.pdf (filename),usctheses-c17-331103 (legacy record id)

Legacy Identifier 9835117.pdf

Dmrecord 331103

Document Type Dissertation

Rights Pobst, Jeffrey Alan

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 au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA