I I I I I I I I!

wil‘lllilﬂlllll‘llﬂl‘ L
LIBRARY
Minn Stat.
University

 

This is to certify that the

dissertation entitled

Experimental Development of a
Microwave Electrothennal Thruster

presented by

Stanley Joseph Whitehair

has been accepted towards fulﬁllment

 

of the requirements for
Ph.D. degree“) Elect. Engr.
G Majorprofessor ‘ ‘
Date "0 We

"(Iii-n. Am—M A ' I" ItL' , . - . 042771

 

 

 

 

 

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

NOR/‘701 3 200

EXPERIMENTAL DEVELOPMENT OF A

MICROWAVE ELECTROTHERMAL THRUSTER

3!!

Stanley Joseph Whitehair

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Electrical Engineering and System Science

1986

ABSTRACT

EXPERIMENTAL DEVELOPMENT

OF A MICROWAVE ELECTROTHERMAL THRUSTER

BY

Stanley Joseph Whitehair

This dissertation reports on the experimental investigation and
application of high pressure microwave plasma discharges in helium.
nitrogen and oxygen. Initial experiments examined the generation and
maintenance of microwave plasma discharges in a cylindrical microwave
cavity at high pressure using different gases. Coupling efficiency,
plasma density and cavity Q versus pressure and gas type were measured
along with other discharge properties. The microwave discharges were
formed separated from any metal electrodes at all pressures and formed
wall stabilized discharges free from contact with any surface at high
pressures. Measured microwave coupling efficiencies into the discharges
were in excess of 958 for all gases, accounting for cavity wall losses
of less then 58. In addition power densities of up to 640 N/cm3 were
measured. .

Further experiments investigated the development and use of a
microwave electrothermal thruster concept. Two different high pressure
microwave coupling devices were developed. The first of these coupling
devices used a coaxial microwave discharge to heat the propellant and a
quartz nozzle to convert it into thrust. The device operated at a

frequency of 2.45 GHz in a TEM mode over a power range of 200 to

n :et-ee
5‘3 W
“3!“: ‘E
”mote
P.

1‘.
cg. ' .‘el

Ir...
Ia. ".“el

600 Watts. Experimental measurements for thrust, specific impulse and
energy efficiency were obtained for nitrogen gas with flow rates of up
to 6000 sccm. Measured thruster energy efficiency varied from 30% to
60*. The second coupling device used a cylindrical cavity microwave
discharge to heat the propellant. The cavity was operated in a TMO11
and TM012 mode at 2.45 0H2 with powers of up to 2000 Watts. Nitrogen
propellant was tested yielding energy efficiencies between 10% to 25%
and a maximum specific impulse of 280 sec. Helium propellant was tested
with power levels of up to 1200 Watts. Measured energy efficiencies
were between 103 and 50% with specific impulses between 200 to 600 sec.
Tests with a metal nozzle produced similar results in nitrogen gas with
efficiencies of up to 20% at a specific impulse of 425 sec. The
performance of these thrusters compared favorably with other
electrothermal thrusters and demonstrated the feasibility of a microwave

electrothermal concept.

DEDICATION

This thesis is dedicated to the memory of Clay and Stan McKeegan

iv

 

ACKNOWLEDGMENTS

The author is thankful for the encouragement and guidance
received from Dr. Jes Asmussen throughout this investigation.
Additional thanks is given to Mr. Shigeo Nakanishi for his help with
experiments presented in this thesis. Thanks is expressed to Mr. Ben
Boyle for his help setting up and running the experiments. Special
thanks to my parent for their encouragement in completing this degree.

This research was supported in part by fellowships from Michigan
State University and by grants from the National Aeronautics and Space

Administrations Lewis Research Center.

n)
.

(A!
.

 

 

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

LISTOF PIGURES........................ ........ .................. ix
LIST OF TABLES ...................................................... xii
NOMENCLATURE............... ......................................... xiii
CHAPTER 1 INTRODUCTION
1 1 INTRODUCTION....... ....... . ............................... 1
1 2 RESEARCH............. ....... . ........... ... ............... 2
1.3 RESEARCH OBJECTIV§§ ....................................... 3
1.4 THESIS OUTLINE .................... . ....................... 4
CHAPTER II MICROWAVE ELECTROTHERMAL THRUSTERS
2.1 INTRODUCTION. .................... ........................ 5
2.2 ELECTRIC PROPULSION ......... ........... ................... 5
2.3 §T§CTROTHERMAL PROPULSION .......... ...... ....... . ......... 1
2.3.1 THRUSTER PERFORMANCE EQUATIONS ................... 9
2.3.2 RESISTOJETS..... ................................. 12
2.3.3 ARCJETS .......................................... 15
2.3. 4 OTHER CONCEPTS ................................... 1s
2 3. 5 SUMMARY ..... .... ............................. 18
2.4 MICR w VE ELECTROTHERMAL THRUSTERS ........................ 19
2.4.1 CONCEPT. . ................ .................. ...20
2.4 2 ADVANTAGES AND DISADVANTAGES ..................... 20
2.4.3 ONBOARD POWER CONDITIONING ....................... 22
2 4 4 BEAMED POWER SUPPLY .............................. 23
CHAPTER III MICROWAVE APPLICATORS AND ARCS
3.1 INTRODUCTION........................ ...................... 25
3.2 MILCROWAVQPPLICATORS ..................................... 25
3.2.1 RESONANT APPLICATORS. ..... . ...................... 26
3.2.2 PROPAGATING WAVE APPLICATORS ..................... 29
3.3 COAXIAL APPLICATORS .............. . ........................ 32
3.3.1 APPLICATOR DESCRIPTION ........................... 32
3.3.2 APPLICATOR OPERATION ...... . ...................... 34
3.4 CAVIIX,TYPE APPLICATORS ........ . .......................... 36
3.4.1 APPLICATOR DESCRIPTION ........................... 35
3.4.2 APPLICATOR OPERATION ............................. 41
3.4.3 APPLICATOR TUNING ................................ 43
3.5 MICROWAVE ARCS ...... .................................. 48
3.5.1 DESIRED ENERGY PATHS ............................. 49
3.5.2 DISCHARGE PROPERTIES ............................. 53
3.5.3 DISCHARGE INITIATION.. ........................... 55
3.5.4 DISCHARGE EXPERIMENTS. ........................... 57

vi

 

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

LIST OF FIGURES................ ...... ... ............................ ix
LIST OF TABLES. ....... . ............. . ........... . ..... ..............xii
NOMENCLATURE................ ...... .............. .................... xiii
CHAPTER 1 INTRODUCTION
1 1 INTRODUCTION....... ....................................... 1
1 2 RESEARCH................... ....... . ....... . ............... 2
1.3 RESEARCH OBJECTIVES.... ................... ....... ......... 3
1.4 THESIS OUTLINE........ ..... . ...... . ....................... 4
CHAPTER II MICROWAVE ELECTROTHERMAL THRUSTERS
2.1 INTRODUCTION. ................... ........................ 5
2. 2 ELECTRIC PROPULSION .......... ............... ............ ..5
2. 3 EL ECTROTHERMAL PROPULSION ......... .......... .............. 7
2.3.1 THRUSTER PERFORMANCE EQUATIONS.......... ...... ...9
2. 3. 2 RESISTOJETS. .................... ................. 12
2.3. 3 ARCJETS. . ..... .......... . ................... ...15
2.3.4 OTHER CONCEPTS......... ..... . .................... 15
2.3.5 SUMMARY ..... ...... ......... . ................. ....18
2.4 MICROWAVE ELECTROTHERMAL THRUSTERS................... .....19
2.4.1 CONCEPT... .............. ......... ........ ....20
2.4.2 ADVANTAGES AND DISADVANTAGES ..................... 20
2.4.3 ONGOARD POWER CONDITIONING ....................... 22
2.4.4 BEAMED POWER SUPPLY........ ...................... 23
CHAPTER III MICRowAVE APPLICATORS AND ARCS
3.1 INTRODUCTION.............. ..... ... ........................ 25
3.2 MICROWAVE APPLICATORS.. ........ . .......................... 25
3.2.1 RESONANT APPLICATORS ..... ....... ............... ..25
3.2.2 PROPAGATING WAVE APPLICATORS ..... . ............ ...29
3.3 COAXIAL_APPLICATORS................. ...................... 32
3.3.1 APPLICATOR DESCRIPTION.... ....................... 32
3.3.2 APPLICATOR OPERATION.. ........... . ..... . ......... 34
3.4 CAVITY TYPE APPLICATORS .......... . ........................ 35
3.4.1 APPLICATOR DESCRIPTION ........................... 35
3.4.2 APPLICATOR OPERATION... .......................... 41
3.4.3 APPLICATOR TUNING ..... ..... ...................... 43
.3.5 MICROWAVE ARCS.................. ......................... .43
3.5.1 DESIRED ENERGY PATHS.... ...... . .................. 49
3.5.2 DISCHARGE PROPERTIES.. ....................... ....53
3.5.3 DISCHARGE INITIATION.... ......................... 55
3.5.4 DISCHARGE EXPERIMENTS.. .......................... 57

Vi

1.1;
T.2j

5.3
5.4

311:? VI
5. 1
5. 2

CHAPTER IV EXPERIMENTAL SYSTEMS AND MEASUREMENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

4.1 INTRODUCTION. ................................... ....... ...50
4. 2 GENERAL EXPERIMENTAL SYSTEM. ............... . ........ 50
4.2.1 GAS FLOW, PRESSURE AND VACUUM MEASUREMENT. ...50
L 2. 2 MICROWAVE SYSTEM. .... ...... ........... ......... 53
L 2. 3 DISCHARGE VOLUME MEASUREMENTS... ..... ...... ....55
4. 3 HIGH PRESSURE CAVITY DISCHARGEvCRITERIA.... ............ ...58
4.3.1 COUPLING EFFICIENCY............... ........... ....59
4.3.2 CAVITY LOADED 0................. ................. 7o
4. 4 THRUSTER PERFORMANCE CRITERIA ............ ........... ..... .72
4.4.1 DIRECT THRUST MEASUREMENT.......... .............. 73
4. 4. 2 POSSIBLE ERRORS IN DIRECT THRUST MEASUREMENTS... .77
4. 4. 3 INDIRECT THRUST MEASUREMENT. . .......... .....80
4. 4. 4 VERIFICATION OF INDIRECT THRUST MEASUREMENTS ..... 81
CHAPTER V HIGH PRESSURE MICROWAVE DISCHARGES
5.1 INTRODUCTION. ................. ........ .. ..... ... .......... 84
5. 2 EXPERIMENTAL SYSTEM. ......... ........ ......... ...........84
5.2.1 MICROWAVE CAVITY TYPE ABSORPTION CHAMBER ......... 85
5. 2.2 MEASUREMENT SYSTEM AND EXPERIMENTAL PROCEDURE....85
5.3 EXPERIMENTAL MEASUREMENTS..... ..... ... ..... ....... ...... ..85
5.4 SUMMARY.................... .......... ...... .............. .103
CHAPTER VI DEMONSTRATION OF A MICROWAVE ELECTROTHERMAL THRUSTER
5.1 INTRODUCTION. ...................... ...... ......... ........ 104
5. 2 EXPERIMENTAL SYSTEM .................. . .............. 104
5.2.1 PROTOTYPE ELECTROTHERMAL MICROwAVE THRUSTER ...... 105
5. 2. 2 MEASUREMENT SYSTEM AND EXPERIMENTAL PROCEDURE....108
5.2.3 COAXIAL THRUSTER DEVELOPMENT ..................... 110
5.3 EXPERIMENTAL MEASUREMENTS ................................. 114
5.4 COMPARISON OF EXEERIMENTAL THRUSTERS......... .......... ...119
CHAPTER VII CAVITY APPLICATOR THRUSTER RESULTS USING QUARTZ NOZZLES
7.1 INTRODUCTION................................... ........... 122
7. 2 EXPERIMENTAL SYSTEM. ..... ..... .............. ......... 123
7. 3 EXPERIMENTAL MEASUREMENTS ........ .. .................... 125
7.3.1 EXPERIMENTS WITH NITROGEN IN DIFFERENT MODES ..... 125
7.3.2 EXPERIMENTS WITH HELIUM IN DIFFERENT MODES ....... 133
7.3.3 DISCUSSION... ............... . .................... 135
CHAPTER VIII EXPERIMENTS WITH METAL NOZZLE THRUSTERS
8.1 INTRODUCTION............................... ........ .......142
8.2 EXPERIMENTAL SYSTEM ..... .... ......... . .................... 142
8.2.1 MICROWAVE APPLICATOR ............................. 143
8.2.2 EXPERIMENTAL APPARATUS... ........................ 149
8.3 EXPERIMENTAL RESULTS ........... . .......................... 151
8.4 DISCUSSION OF RESUEIE ..................................... 159

 

Vii

CHAPTER IX CONCLUSIONS

 

 

 

 

 

 

9.1 SUMMARY OF RESULTS ......... . ............................. .160
9.2 CONCLUSIONS... .............. . ............................. 162
9.3 RECOMMENDATIONS ........................................... 162
APPENOEX A
A.1 INTRODUCTION..................... ............. . ........... 165
A.2 EXPERIMENTAL REVIEW............................... ..... ...169
A.3 MICRONAV§:§OM§USTION FLAME RESEARCH... .................... 178
A.‘ CONCLUSIONS. 0 O C C C I O O I I O I O l O O O O O O O O O O I I O O I O ........ O ....... 186
REFERENCES...‘......OOOOOCOOOOOOO. OOOOOOOOOOOOOOOOOOOOOO .... ......... 187

viii

LIST OF FIGURES

Figure 2.1 Electrothermal Thruster........................ ......... 8

Figure 2.2 Typical Resistojet Thruster.............................13
Figure 2.3 Comparison of Different Electrothermal Thrusters........14
Figure 2.4 Typical Arcjet Thruster......................... ..... ...16
Figure 2.5 Microwave Electrothermal Thruster Concept...............21
Figure 3.1 Coaxial Reentrant Cavity................................28
Figure 3.2 Radiating Have Applicator...............................31
Figure 3.3 Coaxial Microwave Applicator............................33
iﬁgure 3.4 Experimental Coaxial Microwave Applicator...............35
Figure 3.5 Cross Section of a Cavity Applicator....................37
iﬁgure 3.6 Picture of a Assembled Microwave Applicator.............39
Figure 3.7 Picture of Disassembled Microwave Cavity Applicator.....40
Figure 3.8 Electric and Magnetic Field Patterns for the

Cavity Applicator.......................................42
iﬁgure 3.9 Equivalent Circuit of the Cavity Applicator.............45
lﬁgure 3.10 Energy Paths Converting Microwave Energy into Thrust....50
iﬁgure 3.11 Microwave Discharge Properties..........................54

Figure 4.1 Gas Flow, Pressure and Vacuum Measurement System ...... ..61
Figure 4.2 Experimental Microwave Supply System ............. . ...... 64
Figure 4.3 Experimental Setup for Discharge Volume Measurement.....67
Figure 4.4 Radial Electric Field for a Nitrogen Discharge .......... 71
Figure 4.5 Front View of Thrust Stand ...... . ...... ............ ..... 74
Figure 4.6 Side View of Thrust Stand ....... .............. .......... 75
Figure 4.7 Thrust Versus Background Tank Pressure..... ........ .....79
iﬁgure 4.8 Comparison of Pressure Ratio to Thrust Ratio..... ....... 83
Figure 5.1 Picture of Discharge Constriction Versus Pressure ....... 87
Figure 5.2 Microwave Coupling Efficiency for Nitrogen..............91
Figure 5.3 Microwave Coupling Efficiency for Helium.......... ...... 93
Figure 5.4 Cavity 0 Versus Pressure for Nitrogen Discharges ........ 94
Figure 5.5 Cavity Q Versus Pressure for Helium Discharges ...... ....95
Figure 5.6 Power Density Versus Pressure for Nitrogen Discharges...96
Figure 5.7 Power Density Versus Pressure for Helium Discharges ..... 97
Figure 5.8 Power Density Versus Pressure for Different Gasses ...... 98
Figure 5.9 Microwave Coupling Efficiency for Oxygen................100
Figure 5.10 Cavity Q Versus Pressure for Oxygen Discharges..........101
Figure 5.11 Power Density Versus Pressure for Oxygen Discharges.....102

ix

4’s 3 ‘5
III Ilv Fl-

9. . A31. Ali ‘ 0A1 uh! I, Ali 9 Q a I
’3 Ali old aha Phi Phi alt Elna pic I 1 O Q. I I a. a a. I a
I'D 0“ O.- «a «a ”II 0.. I.- ”Io I'D ‘ I. “U! I‘- l‘a 0.. M.- ”I‘ “I. "I On.-
an. at. al. 0*. GO- «a. or a“. at. in. c n. Eh. I u. o u: u u. an a n «nun MN. .0": RW-

93.5
":3'! 5.6

Figure
Figure
Figure
Figure
Figure
Figure

Figure
Figure
Figure

Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure

ciao;
0.0.0. 0.. 0
COO-d 0501‘de

(peanut-uni

QQQQQQQQ

Experimental Coaxial Microwave Thruster.... ............. 106
Enlargement of Discharge Chamber........................107
Experimental Setup of Microwave Thruster...... ........ ..109
Experimental Microwave Thruster Design Variations.......111
Microwave Thruster Design Variations... ....... ..........113
Picture of 8 Operating Microwave Electrothermal

Thruster................................... .. ...... .115
Coaxial Microwave Electrothermal Thruster Performance...116
Power to Thrust Ratio for the Microwave Thruster........118
Electrothermal Thruster Comparison......................120

Cross Section of a Cavity Microwave Thruster............124
Energy Efficiency Versus Input Power for Nitrogen.......129
Results of Experiments by S. Nakanishi..................131
Energy Efficiency Versus Specific Impulse for Nitrogen..132
Energy Efficiency VersUs Input Power for Helium.........135
Energy Efficiency Versus Specific Impulse for Helium....137
Energy Efficiency Versus Specific Impulse Comparison....139
Specific Impulse Versus Thrust to Power Ratio

Comparison.. ....... ............. ......... .... ........... 140
Cross SeCtion Of a Microwave Thruster with a

Metal Nozzle ....... .. ..... . ........ .....................144
Enlargement of Cavity Base Plate .......... ...... ........ 145

Microwave Thruster with a Spherical Discharge Chamber...147
Microwave Thruster used for the Experiments in this

Chapter.... ..... .. ............ ......... . ............ .148
Experimental Setup of Microwave Thruster ................ 150
Input Power Versus Specific Impulse.......... ........... 153
Energy Efficiency Versus Specific Impulse ............... 154
Specific Impulse Versus Thrust to Power Ratio... ........ 155
Energy Efficiency Versus Input Power Comparison ......... 156

Energy Efficiency Versus Specific Impulse Comparison....157
Specific Impulse Versus Thrust to Power Ratio

comparisonOOOOOO......OOIOOOOOOOOOOOOO0......0... ....... 158

Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure

Figure
Figure
Figure

dCDOQOM4hUN—a

0

>>> ?>?)>>>>>>

O
add-I
con-s

Diffusion and Premixed Flames ........................... 167

Flame Zones. . ...................... . ..... ... ...... . .168
Experimental Setup of Jaggers and Von Engel ............. 170
Floating Flame Apparatus of Jaggers and Von Engel ....... 172
Experimental Setup of Tewari and Wilson ................. 174
Experimental Setup of MacLatchy, Clements and Smy ....... 175
Experimental Setup of Clements, Smith and Smy...........177
Experimental Setup of Hard .............................. 179
Cross Section of Flame Experiment ....................... 180
Cross Section of 8 Flame Experiment with a

Bunsen Burner............................. ........ . ..... 182
Experimental Microwave Sweep System ..................... 183
Experimental Results .................................... 184
Experimental Results .................................... 185

Xi

QII ﬂirt At...
0‘! Old Oh I
I'- .‘- a.-
.II 9:. CI-

i-
9|.-

l.‘
I

‘
I

LIST OF TABLES

Table 5.1 Experimental Results in 12mm Discharge Chambers .......... 88
Table 5.2 Experimental Results in 25mm Discharge Chambers ......... .89
Table 5.3 Experimental Results in 37mm Discharge Chambers .......... 90
Table 7.1 Operating Conditions for Experiments with Nitrogen
Propellant....... ...... .... ...... .. ..................... 128
Table 7.2 Operating Conditions for Experiments with Helium
Propellant ............................................... 134

Table 8.1 Operating Conditions for Experiments Presented in
Chapter 8 ................................................ 152

xii

spc
spH
J'BL
jx

jxin

Nomenclature

Capactance of Excited Mode Near Resonance
Electron Charge

Overall Microwave Coupling Efficiency
Applicator Microwave Coupling Efficiency
Electric Field

Radial Electric Field

Empty Cavity Radial Electric Field
Permittivity of Free Space

Permeability of Free Space

Thrust Force

Cold Propellant Thrust Force

Hot Propellant Thrust Force

Gravitational Acceleration (9.8m/sz)
Conductance of Excited Mode Near Resonance
Gigahertz

Conductance of Discharge

Magnetic Field

Total Input current on the Coupling Probe
Specific Impulse

Cold Specific Impulse

Hot Specific Impulse

Susceptance of Discharge

Reactance of Modes Away From Resonance
Cavity Input Reactance

Kilowatt

Inductance of Excited Mode Near Resonance
Tuning Probe Length

Tuning Short Length

Megahertz

xiii

eagoegzolfi

«F 8°

P/F

<p>

Rin

TE
TEM

”eff

Mass Flow

Mass Flow of Hot Propellant

Mass Flow of Cold Propellant
Molecular Weight

Transformer Turns Ratio

Mass Electron

Electron Density

Thruster Efficiency

Excitation Frequency

Plasma Frequency

Cold Chamber Pressure

Hot Chamber Pressure

Power to Thrust Ratio

Power Density

Power Density per Electron

Power Absorbed in the Discharge
Power Absorbed in Cavity Halls
Incident Power

Reflected Power

Total Power Absorbed in the Microwave Cavity
Empty Cavity Absorbed Power
Cavity Q

Loaded Cavity Q

Empty Cavity Q

Intrinsic Resistance of the Cavity Walls
Cavity Input Resistance

Interior Cavity Hall Surface
Transverse Electric Mode
Transverse Electric Magnetic Mode
Transverse Magnetic Mode

Gas Temperature Prior to Expansion

Collision Frequency

xiv

Exhaust Velocity of Cold Propellant

Exhaust Velocity of Hot Propellant

Exhaust Velocity of PrOpellant

Cavity Volume

Plasma Volume

Time Average Magnetic Energy Stored in the Electric
Field

Time Average Magnetic Energy Stored in the Magnetic
Field

Cavity Input Impedance

Transmission Line Characteristic Impedance

XV

.J

CHAPTER I

INTRODUCTION

1.1 INTRODUCTION

This dissertation is concerned with the experimental investigation
and application of high pressure microwave plasma discharges in helium,
nitrogen and other gases. Initial experiments studied the generation
and maintenance of microwave plasma discharges in a cylindrical
microwave cavity at high pressure in different gases with no gas flow.
Coupling efficiency vs. pressure and gas type and other discharge
properties are measured. Further experiments deal with the development
and testing of a microwave electrothermal thruster concept. This
microwave electrothermal thruster concept is the application of a high
pressure microwave plasma discharge to heat a gaseous propellant to a
high temperature, and then using a nozzle, convert the thermal energy of
the propellant into thrust.

Experimental research developed two different coupling devices
amich produce the desired high pressure microwave plasma discharges.
The first of these devices is a coaxial microwave applicator and the
second is a cylindrical microwave applicator. Thrusters using both
coupling devices were used to measure thrust efficiency and other
properties for several different gases.

This thesis summarizes the experimental research and development
that lead to a successful demonstration of a working microwave

electrothermal thruster. Some of this work has already been published

Y‘

'51

vi

a.“

I-

in scientific noblications1'2, at international scientific
conferences3"'5'5'7'a, and in patent disclosuresg'IO. The experimental
devices that were developed, represent an early prototype of an
electrothermal thruster engine concept that has the potential for
improvement and further study. These prototype thrusters were designed
primarily to lend themselves to the experimental and theoretical
investigation of electromagnetic-plasma interactions, microwave

discharge energy balance, emission spectroscopy, etc..

1.2 RESEARCH‘

The research described in this thesis was carried out by the
author over the period of 1982 to 1986 at Michigan State University
under the direction of Dr. Jes Asmussen. The experimental work
described in this thesis (except Chapter 6) took place primarily at the
Michigan State University Plasma Research Laboratories. Mr. Shigeo
Nakanishi of NASA Lewis Research Center provided additional aid in the
design, operation and testing of the actual electrothermal thruster.
Experiments with the initial coaxial microwave electrothermal thruster
were carried out at Lewis Research Center using vacuum tank #8 under the
direction of Mr. Nakanishi and are presented in Chapter 6. These
experiments were carried out at Lewis Research Center because of the
availability of a low pressure vacuum tank and thrust stand allowing
direct thrust measurement.

The research described in this thesis builds directly on previous
research carried out by Dr. Fredericks“, Dr. Mallavarpu12 and Dr.

3

Rogers1 The success of this thesis is dependent upon these earlier

experimental and theoretical studies of microwave discharges inside

 

‘ixil

I I... I

IE-
'
an!
__.

3

microwave cavities. Indeed the theoretical knowledge used to build and
understand the cavity and coaxial applicators were developed by these
earlier experiments. The thesis research of Dr. Rogers, concerned with
the properties of high pressure argon discharges, formed much of the
background for the cavity research presented in this thesis. Further
the basic microwave applicator the coaxial thruster is based on, was

designed and built by Dr. Rogers.

1.3 RESEARCH OBJECTIVES

The objectives of this thesis research are primarily experimental.
The main objective in the first experiments were to: (1) generate and
maintain high pressure microwave plasma discharges in several different
gases in a no flow situation, (2) take measurements of coupling
efficiency versus pressure, and (3) take additional measurements of
other plasma properties such as power density and cavity Q versus
pressure etc.. The main objective in the second set of experiments were
to: (1) generate and maintain high pressure microwave discharges in
several different gases in a gas flow situation with different types of
microwave couplers, (2) using this information produce a working
microwave electrothermal thruster based on each type of microwave
coupler studied earlier, (3) take measurements of these thrusters such
as energy efficiency and specific impulse, and (4) conduct experiments
to improve these thrusters by improving their performance
characteristics such as increasing specific impulse and energy
efficiency.

While the main topic of this thesis is microwave generated plasma

discharges, a related experiment is presented in the appendix. This

 

4

experiment involves the measuring of coupling between a microwave field
in a cylindrical cavity, to a plasma flame generated by the burning of a
gas. The purpose of this experiment is to show that energy in the form
of microwaves can be coupled into a conventional combustion flame. The
interest in doing this is that the coupled energy may be able to change

the flame characteristics in a beneficial manner.

1.4 THESIS OUTLINE

This dissertation is organized as follows. Chapter II presents a
background and review of previous work on electrothermal thrusters and
provides a introduction to the concept of a microwave electrothermal
thruster. Chapter III presents a brief review of previous work on high
pressure discharges and describes the microwave applicators used in
these experiments to couple energy into the discharge. Chapter IV
describes the basic experimental setup used to conduct the experiments
and measure data. Chapter V presents an experimental demonstration of
coupling to high pressure microwave discharges and the measurements of
the properties of these discharges. Chapter VI contains the
experimental demonstration of a working microwave electrothermal
thruster and its measured performance characteristics. Chapter VII
presents experiments with cavity type applicator thrusters. Chapter
VIII presents experiments with metal nozzle type thrusters. Chapter IX
presents the conclusions and some speculation on improvements that could

be made in thruster design.

 

CHAPTER II

MICROWAVE ELECTROTHERMAL THRUSTERS

2.1 INTRODUCTION

This chapter presents an introduction to the concept of a
microwave electrothermal thruster. The first section of this chapter
reviews previous work in the area of electric propulsion, specifically
in the area of electrothermal thrusters. A short introduction to
electric propulsion is presented along with a description of
electrothermal thrusters and their application. Two main types of
electrothermal thrusters, the resistojet and the arcjet, are currently
being considered for use in space. A review of these two concepts is
presented along with other experimental concepts. A summary of the
performance of these thrusters is presented. In Chapter 6 this summary
vﬁll allow a comparison between earlier electrothermal work and the
microwave electrothermal thruster experimental data presented in this
thesis. The second section of this chapter gives an introduction to the

concept of a microwave electrothermal thrusters.

2.2 ELECTRIC PROPULSION

Chemical, electrical and nuclear are the three principle types of
spacecraft propulsion under study today. Chemical propulsion“, which
is by far the oldest, best studied and best developed type of

propulsion, provides large amounts of thrust necessary in missions such

 

 

as those that have to overcome gravity. However for long-term missions,
a chemical engine's rapid use of fuel limits its usefulness. Nuclear
propulsion‘s, by conversion of mass into energy has the potential of
providing large amounts of thrust on a long term basis. However nuclear
propulsion is a new concept, still under development and with potential
questions of safety and cost not yet answered.

Electrical propulsion, in its different forms, uses electricity to
increase the exit velocity of propellant. If this exit velocity can be
raised to a higher value than possible in a chemical engine, then a
electric engine could provide the same amount of thrust while using up
less propellant in the process.

There are three main types of electric propulsion, electrostatic,
electromagnetic and electrothermal. They differ in the method that is
Lmed to convert the electric power into thrust. An electrostatic
ﬂwuster uses an electric field to accelerate ions to produce thrust.

It is typically referred to as an ion thruster. An electromagnetic
‘ﬂwuster uses a magnetic field interaction to convert electric power
into thrust. Two types of electromagnetic thrusters are the MPD
ﬂwusters and the pulsed inductive thrusters. An electrothermal

‘ﬂwuster uses electric heating to convert electrical power into thrust
\Ha a nozzle. An additional concept for electric prOpulsion is the free
radical concept15'17 using dissociated hydrogen to produce thrust. All
ﬁwther discussion of electric propulsion will be limited to
eiectrothermal thrusters since the thruster concept described in this
thesis uses this energy transfer method.

While electric propulsion is not as well developed as chemical

engines, substantial research and development has been done on

 

1‘

electrical engines over the last thirty years. Jahn18 gives an

excellent description of basic principles of electrical propulsion, its
advantages and disadvantages, and examples of working electrical
propulsion concepts. He also provides descriptions for the many
different types of electrical prOpulsion that are currently being
studied. Because of this, this thesis will provide descriptions of only
those concepts that most closely relate to the microwave electrothermal

thrusters and the reader should refer to Jahn for more information on

other concepts .

2. 3 ELECTROTHERMAL PROPULSION

 

A diagram of a generic electrothermal thruster is shown in Figure
2'1' A PrOpellant that could be a gas, liquid or a solid in some
t"a'liiPOPtable formis fed into a thermal absorption region. In this
"Qion the propellant is heated to a higher temperature. The propellant
is then exhausted through a converging-diverging nozzle that converts
“‘9 “‘9“ temperature propellant into thrust. The greater the
temeratur‘e increase in the thermal absorption region, the better the
Operation of the thruster. However limitations in materials restrict
Operation of such thrusters to certain design configuration to produce
”HUNG Electrothermal thrusters and will be discussed in the following
sections.

For a given mission, the type of thruster that one would select
"WM be based on a great number of considerations. While the
Nrformance of the thruster alone would an important factor, it would
not be the only factor. Other factors such as weight, volume, power

'- .
qurement, propellant requirement, etc., must be considered also. The

Electrothermal
Thruster

 

 

 

0/ \
///////

 

 

 

 

Thrust
iOOl—> /erI/ H ”Ix/J
P /Absorption/ Proepil ffff _9

 

gm“: // Region/ W
////////////
///////////

L /

 

 

 

 

 

 

Converging-
Diverging
Nozzle

 

.41

actual decision would be based on a combination of all of these factors.
Because of all the factors that could be considered, this thesis will
restrict the performance discussion and comparison to primarily the
performance of just the thruster, while some discussion of the general
thruster system will be given. Primarily this means that the criteria
of performance for this thesis will be based on several numerical
quantities derived from experimental measurement of actual thruster
Performance and will be discussed in section‘2.3.1.

The most developed types of electrothermal thrusters are mainly
the resisto jet and the arc jet. Both of these concepts have potential
aPPlications in space missions. Sections 2.3.2 and 2.2.3 describe in
'0?! detail advantages and disadvantages of each of these types of
t"Puriters. The last section presents descriptions of some other
Potential types of electrothermal propulsion. This section also

provides a summary of results for latter comparison to other thrusters.

2.3.1 THRUSTER PERFORMANCE EQUATIONS

The evaluation of thruster performance involved the experimental
determination and calculation of several figures of merit. These
“WWW“: are generally recognized as the standards for thruster
performance comparison and can be found elsewhere, but because of
Variation in the definitions of these quantities, especially energy
efficmmv. the definitions used in this thesis are briefly outlined in
this section. The first of these is thrust force, F, and is defined as
bel'ng equal to exhaust mass flow rate times the exhaust velocity
"Native to the thruster. Thrust is measured directly or calculated

fr“ other measured quantities. The overall energy efficiency, 72, the

 

10

qnmifk:impulse. Isp' and the power to thrust ratio in are described in
detail in this section.
The specific impulse is defined as the ratio of the thrust to the

propellant mass flow expressed in units of seconds. Thus,

Isp' F . (2-1)

*9

where ma mass flow rate of propellant
g- gravitational acceleration (9.8 m/sz)
F- thrust force
Overall energy efficiency is defined as the time rate with which
kinetic energy is imparted to the hot propellant exhaust divided by the
sum of the input power and the rate of kinetic energy flow of the
initially cold propellant. Thus,

1 2
3"HVH

2? = (2.2)

where a“. mass flow rate of hot propellant,
VH' exhaust velocity of heated propellant,
Pas microwave power absorbed by the discharge,
Vc' cold gas exhaust velocity,
mc-mass flow rate of cold gas.

"0" momentum considerations, the thrust force is given by F=mv,

assuming all mass particles have a uniform velocity, vp, parallel to the

 

err

ll

thrust vector but in the opposite direction. Thus,

,7 a _ (2.3)

where EH: hot condition thrust,
Fcz cold condition thrust.
Note that when the input absorbed power is zero, output conditions
CO'PGSPOnd to the cold flow input, yielding an efficiency of 100
percent. In the experimental results presented in this thesis, the mass
flow rate was held constant during the hot and cold measurements, so

that P' ﬂic- it". Thus, from equations 2.1 and 2.3, the efficiency can be

2
Ispc

expressed as ,

 

 

 

12spH
7) = 2 = (2.4)
2Pa +(Ispc) 2Pa + 1
92m gzm(15pc)2

This definition of energy efficiency is not universal, some definitions
do not consider the cold gas kinetic energy and some definitions do not
include all the input power in the definition of input energy.

The Power to thrust ratio is simply the total input power divided
by the thrust force. Using the definition of energy efficiency 9W9"
abWe. the power to thrust ratio can be written,

I (I )2
spH _ spc (2.5)

n IspH

TlI'O
I
NI“)

12

2.3.2 RESISTOJETS

Figure 2.2 shows the basic operating principle of a resistojet
thruster. In this type of electrothermal thruster the thermal
absorption region is produced by resistive heating of a electric
element. This heating raises the temperature of the propellant passing
through the region. The propellant then passes through a nozzle
producing thrust.

Different types of resisto jets have been built and tested. They
differ mainly in the flow configuration used to heat the gas, the amount
0* Power they can handle and the type of propellant they use. Several
9"Miles of experimental resisto jets are presented.

One example is of a 3 kW concentric tubular resistojet.19 The
"'9'" Probellant is fed through a series of concentric tubular
resistance heating elements heating the gas and minimizing the heat loss
fI‘OI conduction and radiation. The thruster was tested with H2 as a
PPOPellant with a input power level of 3 kW. It produced a maximum
Specific impulse of 830 sec with a mass flow of 79x10'6 kg/s and had a
b9“ ”Stem efficiency of 79*. In continued experiments with this type
0f thruster-2°, the energy efficiency vs. specific impulse are presented
in Figure 2.3 as the points designated A, for H2 and NH3 propellant.

A Second example of a working resistojet is referred to as a
HiPEHT (High Performance Electrothermal Hydrazine Thruster) type
device.21 The HiPEHT is a hybrid type device that combines a chemical
°Dgine with an electrothermal augmentation device to increase the
8Pacific impulse of the engine. When tested as a resistojet the
°h°ﬂcﬂ Portion of the thruster was not used and the propellant was

directly added into the electrothermal stage. This electrothermal stage

l3

Resistajet
Thruster

 

 

 

 

 

 

 

 

 

\
I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Heating Element

 

Figure 2.2 Typical Realstojet Thruster

 

'W//////////]
A \LV///////\//ﬂ——
ﬁ///////////'J Th r U S t
: > r////////\///}— —l
000' > Heated —>
Pr °P°"0nt Propellant "—9
— \ 9 I} —>
— 7 [/////f/////}
l —[////////\//A )(
V @//////////.L
—V///////\///l
L
C a n ve r g in g —
Thermal Electric Diverging

Nozzle

14

Comparison of Several
Electrothermal Thrusters
Operating Below 5kW

 

 

 

A,B - Resistolet C - Arcjet
Electrothermal Electrothermal
Thrusters Thruster
100
Propellant
«.1 LBS <_ "2N
no a O-NHJ
N 30 " Ems} 0 -He
1 4 an] 0-H
>5 3%
eo—é \°§b_\_9_ 0—4.9

l

w- MW

@222

1

Energy Efficienc
8 3
1 L

.004 0%

O I I I i I 1 I
O 200 400 600 800

 

 

 

Specific Impulse - Seconds

Thrust in Newtons
Electrothermal .652!
ml:

Thruster
Propellant Flow Rate in 10-5 Kg/s

Figure 2.3 Comparison of Different Electrothermal Thrusters

all
r!

"i

(a

mew

Ni...

 

.m

15

used a vortex heat exchanger to increase the temperature of the

PPOPGHIM- N2,H2 and NH3 gases were tested as propellant. Typical

results for Hz were Isp of 550 sec with 5 kW of input power and an

overall efficiency of 61*. Results for all three gases are presented in

Figure 2.3 as points with the designator B.
A resistojet was the first electrothermal thruster to be tested on

an actual spacecraft in orbit.22 This was a small device using only

90 Watts of power, and with N2 as propellant, generated a specific

impulse of 123 sec. The device was use on board a Vela satellite to

 

adjust the orbital position.
These experiments indicate that resistojets are valid

electrothermal thrusters and have been used in real spacecraft

applications. They posses several advantages for their use in an actual

aplblications. They are easy to start, can use a broad range of power

in”tats in terms of both voltage and frequency, are simple to operate and

can use a wide variety of propellants, although 02 and H20 could present

"ifetime problems. However they suffer from a temperature limitation in

“terials used to heat the gas, limiting the maximum specific impulse to

t"New. 1000 sec.

243.3 ARCJETS
Figure 2.4 shows the basic operating principle of a arcjet

th"taster.” The thermal absorption region in this thruster is produced

using a dc electrical arc discharge. In this manner the the propellant

f‘owing through the discharge region is heated by the arc discharge and

then exhausted through a nozzle producing thrust.
An example of a working arcjet is given by John.24 This thruster

 

16

Arcjet
Thruster

 

 

in
Thrust1

 

 

 

 

 

 

r

 

 

 

PropeHant
\ —>

 

Cathode _.
//ﬂ Anode-

PropeHant

 

 

 

Wall Stabilized Constricted
DC Discharge

Figure 2.4 Typical Arcjet Thruster

 

17

is referred to as a radiation cooled arcjet engine having a constricted
arc configuration. The constricted arc configuration means that a
Laminar column arc is formed so that it passes through the narrowest
portion of the nozzle and it uses the constriction to maintain arc
stability through thermal energy interactions with the constriction
walls, i.e. a wall stabilized arc is formed. This engine was tested in
both H2 and NH3 propellants and and at power levels of 30 and 215 kW.
For 30 kW of input with H2 as a propellant a specific impulse of
1550 sec with a mass flow rate of 100x10”6 kg/s, and a efficiency of 38%
was reported. For 215 kW with Hz a specific impulse of 2200 sec with a
mass flow rate of 330x10"6 kg/s, and a efficiency of 362 was reported.

A one month lifetime of useful operation is reported due to cathode mass

loss.

McCaughey25 presents results for a 1 kW arcjet thruster. This
thruster used both a short constriction and a magnetic field to maintain
arc stability. The thruster was tested in a wide variety of gases
i"<-='|uding H2, He, N2, argon, ammonia, methane, air, etc.. Results for
"2 30d He are shown in Figure 2.3 for this thruster and are designated
as C. Lifetime results of these experiments were reported as good for
p"Oi-'Dellants not containing oxygen or carbon. Use of propellants
°°ntaining oxygen and carbon led to severe material problems.

Resistojets have limited specific impulse, but as can be seen from
these examles, arcjets have the capability for large specific impulse
since the arc can produce much higher temperatures. Arc jets on the

°ther hand have limited lifetime problems and suffer from low
Q‘Fl‘iciencies. In terms of power handling, arc jets of over 200 kW have

tNan built and tested.26 At these power levels the efficiency of the

in.

M
"a

Its

fl.‘

18

are jet imroves from that reported to over 402.27

2.3.4 OTHER CONCEPTS

The resistojet and arcjet are the two best developed
electrothermal concepts. A third concept would be to use high power
laser energy from a remote location to power a thruster.28 The laser
power could be beamed from earth or from another satellite in earth
orbit eliminating the need for a heavy onboard power source. This laser

thermal thruster engine would operate in a pulsed mode or continuous

 

wave mode where laser energy is focused into a small chamber heating a

gaseous, liquid or even solid propellant to many thousands of degrees
then expanding it out a nozzle to produce thrust. At this time it has
not been experimentally demonstrated, but stable laser supported plasmas
have been demonstrated in the laboratory29'3o and the question of
Stability and energy coupling are currently being studied31'32.

The main problem with the arcjet thruster is the arc-cathode-anode
interaction results in both lower efficiency and in erosion. One
hYPOthetical concept that could be used to eliminate this problem would
be to develop a induction arc jet thruster operating at RF frequencies.
such a thruster would not have electrodes in the discharge zone and thus
right be able to combine the advantages of the resistojet and arcjet.
considerable experimental work has been done on induction arcs33'34 and

industrial applications have been studied”.

2 ~3.s sumv
The results summarized above indicate that electrothermal

tl'Irusters operate at high specific impulse when compared to chemical

I a
a
fa.

 

.«i ..

of.‘

 

19

thrusters, and operate at higher levels of thrust than current ion

thrusters. Recent studies have shown that for near Earth orbit transfer

missions, the electrothermal thrusters show the best trip time
performance values of specific impulse of from 1000 to 5000 sec.36

These advantages have resulted in the identification of electrothermal
thrusters as a technology for satellite station keeping and as auxiliary
propulsion of large platforms in low earth orbit.37 Specifically a
thruster with the efficiencies of the resistojet but able to operate at
higher specific impulse (over 1000 seconds) and higher thrust output
would be desirable.38 The ability to use waste products from space
station operations, containing 02 and other chemically active gases, as

propellant39 without the decreasing the thruster lifetimes would also be

highly desirable.

 

2.4 wrcnowaws ELECTROTHERMAL THRUSTERS

From the discussion in the previous section it can be concluded
that there is a need for an electrothermal thruster capable of operating

at specific impulse levels above 1000 seconds with higher efficiencies

and longer operational lifetimes. This section presents the microwave

e1ectrothermal thruster concept which has the potential to overcome
these problems. The basic concept is first presented, next its
at"tantages and disadvantages, a discussion of the power supply is then

”haented and last the potential for beamed power supply is discussed.

 

DJ

«'1'

..f

 

20

2.4.1 CONCEPT

The principal elements of a microwave electrothermal thruster
system are shown in Figure 2.5. The system receives its electrical
energy from a source such as solar cells and converts it into microwave
power via a power conditioner such as a magnetron, klystron or gyrotron.
Once converted into "microwaves", the electric energy is coupled into an
energy absorption chamber where a low molecular weight gaseous
propellant is heated as it flows through the chamber. The heated,

propellant then exits via a conventional nozzle producing thrust.

2.4.2 ADVANTAGES AND DISADVANTAGES
The principle advantage of a microwave electrothermal thruster is

that a microwave discharge does not require the electrodes of a DC
arcjet to operate. This provides several potential advantages for this
concept over the resisto jet and the arc jet. Since there are no hot
electrodes in contact with the discharge, chemically active gases such
as 02 can be used as propellant. A second benefit of this is that there
is no erosion of the electrodes with the potential of longer thruster
Lifetime. If the discharge can be formed completely away from all solid
Co'ﬁponents of the thruster, there will be fewer problems with material
lifetimes and the energy losses from the discharge will be reduced. In
add“ition, if the discharge can be maintained away from solid materials
i“ the thruster, it could be operated at higher temperature than
0t"Ierwise possible, thus increasing the rate of energy transfer to the
propellant. Elimination of the electrodes also allows new freedom in

thi‘uster design .

The principle disadvantage of the concept of a microwave

 

21

Microwave Electroth ermal

Thruster Concept

Beamed Microwave or
Millimeter Wave Power

 

 

 

 

 

 

 

 

 

 

 

 
 
  
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Energy (_' ULTgclt
Storage }; ‘ 603,0");
FT W Propellant
‘ Energy "—"—:
Energy Power ...... _ :
LSource F Conditioner .22 3:23:53?" 1;
‘ l T Nogzle
Propellant
Storage
S p a c e c ra ft

Figure 2.5 Microwave Electrothermal Thruster Concept

 

1‘

.‘1

5.}

9?

22

electrothermal thruster is that it is highly experimental, and at this

time, still under development with other potential problems not even

identified. A further potential advantage or disadvantage is that the

energy absorption chamber requires microwave frequency power to operate
If microwave power is available for other uses in a spacecraft this is
advantage, but if it is not the total thruster system (including the
microwave power conditioner) must be compared to other thruster systems
to determine if it is a potential advantage or disadvantage.

This thesis deals with the development of the microwave

 

electrotheer thruster, however a short discussion of the problem of
the microwave power conditioning will be given in the next two sections.

Two different microwave energy supply systems are described in these

sections to provide this power. The first system describes the

necessary components to generate the microwave power on board and the

second system would function on power beamed to it from a external
Source.

2.4.3 ONBOARD rowan CONDITIONING

Electrical power would be provided by a onboard source of

e1°<=tricity such as solar cells, fuel cells, nuclear reactor, solar

theI‘mal reactor, etc.. At this time the only high power source of

.‘ici‘owave power are tube devices that require high voltages. Thus the

Q"ectrical power would have to be raised in voltage to supply power to
those tubes implying power loss in the conversion.
Tube devices have existed for many years and are well studied and

“Tliierstood.‘°"1 Tube efficiencies of up to 83% with power levels of up

‘20 25 kW at 2.45 GHz have been reported for magnetrons.42 Experimental

23

amplitrons have reported efficiencies of up to 908 with power levels of
up to 400 kW.43 Depressed collector klystrons have reported
efficiencies of over 803 with power levels of up to 1 MW.44

Thus, in any microwave electrothermal thruster system that uses a
onboard power source, the efficiency of the entire system, not just the
thruster, must be considered. If the high voltage power or even the
microwave power is available on board for other purposes, the use of a
microwave thruster may be advantageous. It should be noted that solid
state devices are being developed that can operate at these frequencies.
These devices currently operate at 2.45 GHz with power levels of up to
100 Watts with direct conversion from low voltage power into microwave
power. At 1.85 GHz, commercial solid state power supplies are available
that can produce power output of up to 1.25 kW continuous and 2.5 kW

units are under development.45

2.4.4 BEAMED POWER SUPPLY

As shown in Figure 2.5 there is the potential for the use of
beamed microwave or millimeter wave electricity to supply energy to the
thruster. Such a system may be practical, probably at very high
microwave or millimeter wave frequencies, where the antenna size/weight
is not too large thereby allowing the beaming of 30-100 GHz energy
thousands of miles. This concept has the potential advantage, similar
to laser thermal thrustersza, of not requiring a heavy onboard power
source and of coupling beamed energy directly into the energy absorption
chamber.

There has been great interest in beaming electricity for solar

power stations to Earth for some time.46 However in these applications

24

low frequency power, 60 hz, is desirable and energy losses result in the
conversion. Beaming to a spacecraft using a microwave electrothermal
thruster may not require conversion and thus could prove to be very
beneficial.‘7 However at this time the beaming of microwave energy is
still experimental and there are many questions regarding its use and

operation.

CHAPTER III

MICROWAVE APPLICATORS AND ARCS

3.1 INTRODUCTION

Conversion of microwaves into high temperature thermal energy
requires the use of a microwave applicator and a microwave plasma
discharge or arc. The applicator serves to couple microwave energy into
the arc and the arc heats the propellant gas. The basic types of
applicators are described in section 3.2. Two different energy coupling
structures or microwave applicators were used in the experiments
presented in this thesis. The first of these is a coaxial type
applicator based on a TEM type electromagnetic mode and is described in
the section 3.3. The second type is a cavity type applicator which uses
a classical type resonant cavity and is described in the section 3.4.
The basic operation of the microwave arc and how it transfers energy to

the propellant is described in section 3.5.

3.2 MICROWAVE APPLICATORS

The practical realization of any electrothermal concept requires
the energy absorption chamber to be designed to perform several
interrelated functions. First, it must efficiently couple electric
energy into a low molecular weight propellant and must heat this
propellant to extremely high temperatures. In addition, the energy

absorption chamber must be designed to prevent significant energy losses

25

 

26

from the heated propellant by radiation and heat conduction and
convection to the chamber walls before it exits the nozzle. These
several interconnected functions must be performed simultaneously within
the energy conversion chamber without imposing chamber wall, electrode,
heater element, nozzle material and life limitations. The final design
of the energy conversion chamber results after the trade off between
these several interrelated functions.

Applicators can be classified into groups by the type of
phenomenon they use to sustain the microwave arc.48 The first are
resonant applicators that use a resonant, or standing wave, field
pattern to sustain the plasma discharge. The second group sustains a
discharge with propagating electromagnetic energy. Thus, in a
propagating wave applicator the discharge is sustained by evanescent
fields or by radiating type fields. Examples of each of these is

presented in the following sections.

3.2.1 RESONANT APPLICATORS

In resonant type applicators two different types of discharges can
be identified, fast wave and slow wave. In fast wave discharges the
phase velocity of the EM wave is faster than the speed of light. In
slow wave discharges the phase velocity of the EM wave is less than the
speed of light.

Examples of resonant applicators that produce fast wave high
pressure discharges are given by Babat49 and Asmussenso. The coupling
structure used in these examples, is formed by a cylindrical resonant
cavity. This cavity is basically a length of waveguide terminated by

two shorting planes set to certain resonant eigenvalue lengths.

I.

27

Microwave power is coupled into the cavity by either a coaxial coupling
structure or by a waveguide aperture coupling structure. At high
pressures (above ~10 Torr) the discharge will fill only part of the
cavity and is separated from all surfaces if properly designed. The
discharge alters the field pattern of the resonant mode thus changing
the resonant frequency of the structure. The structure can be retuned
by altering the position of one of the shorting planes51 or by changing
the input frequency of the microwave energysz. This process is
nonlinear since changing the cavity tuning changes the plasma properties
requiring the cavity to be further retuned, etc..

There are several advantages for using this type of energy
applicator in a thruster system. Because it is a resonant device, the
discharge it produces is stable in position irregardless of the rate of
propellant flow, and if properly excited, this discharge can be
maintained away from the chamber walls. The input power level this
device could be maintained at is limited mainly by the input coupling
structure and not by the applicator. An additional advantage of this
applicator is that it can also produce slow wave discharges. Its main
disadvantage is that its structural size relates directly to the
excitation frequency. At higher microwave frequencies its reduction in
size limits the input power it can handle and at lower frequencies its
size may become to bulky for practical use.

An example of a resonant applicator that produces a slow wave high
pressure discharge is given by experiments with the surface wave
13“"Ch9' ShOW" 1" Figure 3-1-53'54 This coupling structure is referred
to as a coaxial reentrant cavity and is operated with a quartz tube

located inside the center conductor. The gap in this structure

 

 

28

Coaxial Reentrant
Cavﬁy

Outer Conductor —___i

Microwave Input
-'” "‘— Gap
Sliding Short 1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

=: 4L

 

 

'30... ‘e 0:0... 0. 0.2.... ’
-‘~:~'.-‘~:~.-'.-'~:~:~‘~

 

 

 

ll :2 A l

1F

 

 

 

 

 

 

 

—— Adjustable Center Conductor
Quartz Discharge Chamber

 

 

 

Microwave Slow Wave Discharge

Figure 3.1 Coaxial Reentrant Cavity

'\
1“.

=1

1‘

(N

29

generates strong fields that launch a surface wave discharge in the gas
inside the quartz tube. This surface wave discharge extends in both
directions from the gap. Energy is transferred to the discharge by the
slow surface wave propagating, thus heating the discharge and
attenuating the surface wave. Increasing the input power to the coupler
has the primary effect of lengthening the discharge.

An advantage of this type of coupler is that it is capable of
operating down to low frequencies (100 MHz) without the need of a large
coupling structure. The coupling gap however presents a problem in
terms of limiting the maximum input power. This is due to possible
breakdown in this region. Another problem is the power handling
abilities of the input coupling structure. The potential for breakdown

limits the maximum value that the input coupling structure can handle.

3.2.2 PROPAGATING WAVE APPLICATORS

Propagating wave applicators can also be divided into two
different types according to how they operate. The first type
presented, uses a evanescent field to sustain a discharge and the second
type presented uses a radiating field.

An example of a coupling structure that uses a evanescent
electromagnetic field is the coaxial microwave "torch".55 A similar
device is presented in great detail in section 3.4 of this chapter. A
microwave discharge is formed attached to a center conductor of a
coaxial matching structure. Gas flow is used to keep the discharge from
arcing to the outer conductor and in the process forces the discharge to
form out straight from the center conductor.

At higher microwave frequencies EM modes other than a TEM mode can

30

form if the diameter of the device is made too large (4 cm for 3 GHz).
But making a smaller device limits the power it can handle without
having unwanted breakdown in the applicator. At lower frequencies this
is not a problem since the diameter can be increased to handle more
power. In fact, since this device uses a TEM mode, it could probably
operate down to low frequencies although changes in the matching circuit
would be necessary. One of the main disadvantages is that the discharge
is attached to the center conductor, possibly causing thermal loss,
melting and limiting the discharge temperature.

An example of a coupling structure using a radiating
electromagnetic field is described in experiments by Moriarty56 and is
shown in Figure 3.2. This structure is basically a quasi-optical
microwave coupling system. This system used a ellipsoidal antenna 10
feet in diameter to focus propagating microwave radiation into a
discharge chamber. A microwave discharge was formed at the focus of
antenna inside the discharge chamber. The system had a coupling
efficiency of 203 with a input peak power of 1 MW at 3 GHz.

For space applications this system is rather bulky but at higher_
frequencies it could be small enough to be practical. In fact it is
similar to the proposed laser thruster28 and would be capable of
operation up to light frequencies as a laser thruster. The matching is
a serious problem and it would have to be improved for real use. The
potential for handling radiation using lenses would allow this coupler

to operate a very high power levels.

31

Radiating Elecromagnetic
Field Applicator

-—-- Pyrex WIndow

—— Discharge Chamber

 

 

 

 

 

Microwave Discharge

 

 

Ellipsoidal Antenna

 

 

Transmitting Horn Antenna

Figure 3.2 Radiating Wave Applicator

ﬁli

32

3-3 W

The basic design and construction of the coaxial applicator
presented in this section was by Rogers57 and was based on earlier
work.55'58 First a detailed description of the basic coupling structure
is given and next a description of the working applicator based on this

coupling structure is presented.

3.3.1 APPLICATOR DESCRIPTION

As shown in the Figure 3.3 the applicator consists of a coaxially
fed microwave discharge located at the end of a 4.8 cm i.d. cylindrical,
brass microwave coupling system. Microwave power is fed in through, the
input microwave port (1), thus entering into a coaxial coupling system
(2) in a transverse electromagnetic mode (TEM). A cylindrical microwave
discharge (3) is formed and maintained at the end of a 2 cm diameter
adjustable center conductor (4). A adjustable sliding short (5) and the
adjustable center conductor allow the coaxial coupler to be tuned so
that all of the incident power is coupled into the discharge structure
(6) located at the end of coaxial applicator. The center conductor of
the applicator is hollow and sealed so that water can flow through it to
cool the tip. Water cooling is also provided for the outside wall of
the coupler although experiments suggest it was not needed. The
discharge structure located at the end of the coupler was designed so it

could be removed and different types of discharge structures could be

tested.

33

 

Coaxial
Microwave
AppHcator

 

 

 

 

Legend -—- 5

 

 

(1)—Coaxial Microwave lnput
(2)-Coaxlal Coupling Structure
(3)—Microwave Discharge
(4)-Center Conductor
(5)—Sliding Short
(6)—Discharge Chamber ”I“

 

 

3:21
D=l

 

 

Figure 3.3 Coaxial Microwave Applicator

34

3.3.2 APPLICATOR OPERATION

Figure 3.4 shows the basic applicator that was tested by Rogers.13
This figure continues the numbering system used in Figure 3.3 with the
following additions. A teflon gas seal (6), seals the discharge chamber
(8) from the atmosphere. The test propellant enters through (7) and
enters into the discharge chamber (8). A microwave screen (9) continues
the outer wall of the coupling system. In these experiments the
applicator was not connected to a vacuum system. The discharge chamber
was sealed and the input gas added to the discharge chamber. A vent
allowed the gas to be removed at the top of the discharge chamber, thus
all experiments were conducted at atmospheric pressure. A piece of
tungsten carbide wire was shorted from the outside conductor to the
inside conductor starting the discharge.

Rogers obtained discharges in helium and nitrogen with a input
power level of 500 Watts and the applicator in a horizontal position.
Reflected power was low indicating good matching. These experiments
demonstrated the use of the torch and the ability to generate high
pressure discharges with it.

Experiments by Rogers with this applicator were continued by this
author. The applicator was changed to a upright orientation and a
0-2.5 kW microwave system was used. The applicator was tested in
mixtures of nitrogen and helium. Short and center conductor tuning
positions were established. The most critical tuning distance was found
to be the distance from the short to the tip of the center conductor.

If this distance was maintained, the short/probe combined action could
be adjusted over a range of values with little change in matching. As

such the torch had no single mode of operation, but rather a range of

35

 

Experimental
Coaxial
Microwave
Applicator E

9—.

 

 

 

 

 

A111

 

Legend

 

(1)—Coaxial Microwave Input
(2)—Coaxial Coupling Structure
(3)—Microwave Discharge
(4)—Center Conductor
(5)—Sliding Short

(6)-Teﬂon Plug

(7)—Gas Input

(8)—Discharge Chamber {H}
(9)-Conducting Screen

 

 

[3:1
D=l

 

 

 

Figure 3.4 Experimental Coaxial Microwave Applicator

gratic
ih
35.1 1
39d an
at m
pacer:
port

“out

tson;
anal i .

ndt

Show
time
is ll

560C

brag

SEC

i911

dis

 

36

operation.

The best match that could be obtained with this applicator was
95‘. In order to simplify construction, unmatched teflon spacers were
used and it was felt that these spacers were limiting the matching that
the applicator could achieve. It was found that the position of these
spacers was very critical. If they were placed close to the sliding
short they would get very hot and also cause arcing between the coaxial

input and the fingers on the short.

3.4 CAVITY TYPE APPLICATORS

The second type of applicator that was used in experiments was a
resonant type cavity applicator. The first section details the basic
applicator design, the next section describes its electrical operation

and the last section describes the circuit tuning of this applicator.

3.4.1 APPLICATOR DESCRIPTION

A diagram of the cavity applicator used in these experiments is
shown in Figure 3.5 and utilizes the same design philosophy of earlier
experiments.51'”-"59 The energy absorption chamber shown in this figure
is made up of two interdependent parts; First, a cavity applicator and
second, a discharge chamber. The resonant ("cavity") portion is formed
by 17.8 cm inside diameter cylindrical brass pipe (1) and transverse
brass shorting planes (or "shorts"). One of the shorts (2) is
adjustable to provide a variable cavity length of 6 to 16 cm. The
second short (3) was fixed in position, but in some experiments could be
removed to allow different discharge chambers (6) to be tested. The

discharge (4), which could be viewed through a copper screen window (5)

37

Cross Section of
Microwave Cavity Applicator

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Legend
(1)-Cavity Walls (7)-Coaxial Microwave input
(2)-Sliding Short (8)-input Coupling Probe
(3)-Base Plate (9)-Brass Collars
(4)-Plasma Discharge (iO)-Microcoax Block
(5)-Viewing Window (il)-Microcaox Probe Hole

(6)-Discharge Chamber (12)-Microcoax Probe

 

Figure 3.5 Cross Section of Cavity Applicator

(5 form
stable

:yi indr
Moe!
‘W |
12 ad
fined

ischa

IicrOI

nlinl
ﬁver
and t
irobe
iiagr
Inuit

leasl

Ippl

disc

inc:
The
3.6
Cav

illll

38

is formed inside a cylindrical discharge chamber and is maintained in a
stable fashion contracted away from the walls of the chamber. The
cylindrical quartz discharge chamber was removable so that different
chambers could be tested. Input microwave power is fed into the coaxial
input port (7) and coupled into the cavity by the adjustable probe (8).
The adjustable probe and the adjustable short allow the cavity to be
tuned so that the maximum amount of power is transferred into the
discharge. Brass collars (9) were used to increase cavity Q and reduce
microwave leakage.

A rectangular brass bar (10) was soldered onto the outside of the
cylindrical outer shell of the cavity parallel to its center axis.
Several small, diagnostic holes (11) were drilled through this piece
and the cavity wall at known axial locations. Small electrical E field
probes (12) made from 2 mm o.d. microcoax were inserted into the
diagnostic hole to measure the radial E field near the wall. The probe
would couple out a small amount of power proportional |Er|2 and then be
measured with a power meter.

Figures 3.6 and 3.7 Show photographs of the microwave cavity
applicator. Figure 3.6 shows the applicator assembled but lacking a
discharge chamber. Figure 3.7 shows the applicator disassembled with
a quartz discharge chamber (0). This discharge chamber has a nozzle
incorporated in it and was used for experiments presented in chapter 7.
The cavity body (A) shown in Figure 3.7 is the same as the one in Figure
3.6 and the microcoax block is visible in the lower right side of the
cavity of both pictures. The same sliding short actuator (B) and the
tuning probe actuator (C) were used in both figures.

Initial experiments with the cavity applicator used a simple type

39

Picture of Assembled
Microwave Cavity Applicator

 

Legend
(A) - Brass Microwave Cavity Body
(note microcoax block in lower corner)
(B) - Sliding Short Actuator
(C) — Tuning Probe Actuator

 

 

Figure 3.6 Picture of Assembled Microwave Cavity Applicator

40

Picture of Disassembled
Microwave Cavity Applicator

 

Legend
(A) - Brass Microwave Cavity Body
(note microcoax block in lower corner)
(B) — Sliding Short Actuator
(C) - Tuning Probe Actuator
(D) - Quartz Discharge Chamber with Nozzle

 

 

Figure 3.7 Picture of Disassembled Microwave Cavity Applicator

:tuator f
not. Th
'ltttl‘ exp
mitjed 3
wt was
'r‘s actua
M to I
actuator 11
"is actua

Icss‘ble h

3.1.2 APPL

in e)
title dis
iiindrica
:Itterns a
3.8. The e
””5 iiiOdt.
1‘! center
ion in F
ﬁmal‘ge
”it” i coc
3' discha

ti discha

41

actuator for the sliding short and no actuator for the variable length
probe. The tuning accuracy of measurement was limited due to this so in
latter experiments the sliding short and variable length probe were
modified and are the actuators shown in Figure3.6 and 3.7. The sliding
short was attached to a jack type actuator as shown in the figures.

This actuator used 3 ACME threaded screws and a coaxial planetary gear
drive to move the sliding short back and forth. A rack and pinion
actuator was added to the variable length probe as shown in the figures.
This actuator allowed the probe to be moved in finer increments than was

possible by hand adjustment.

3.4.2 APPLICATOR OPERATION

In experiments a microwave discharge (4) was created in the center
of the discharge chamber (6) by exciting the cavity in the single, TM012
cylindrical cavity mode (Ls - 14.4 cm). The electric and magnetic field
patterns and the associated discharge for this mode are shown in Figure
3.8. The electric field has an axial standing wave maximum along the
axis producing an intense, approximately half wavelength discharge in
the center of the cavity and the center of the discharge chamber. As
shown in Figure 3.8, the discharge is "contracted" or separated from the
discharge chamber walls. Thus, the discharge has a hot central core
with a cooler gaseous outer layer adjacent to the quartz tube walls. If
the discharge is allowed to touch the quartz walls heat transfer from
the discharge to the walls increases resulting in wall melting.

Experiments described in this thesis also used the TM011 cavity
mode to create a plasma discharge, which is also shown in Figure 3.8.

The electric and magnetic field patterns are similar to the fields of

 

 

/.\r.\. _

/\

L L

I Ch \

 

42

Electromagnetic Field Pattern

for the

Microwave Cavity Applicator

CROSS SECDON
OF QUARJZ TUBE

V) DISCHARGE

<;‘_L :Vﬂume=vo

 

 

O

V

QUARTZ
0/ N, QUARTZ
i

 

 

g \/H —1

\\\TE
CAVITY
(Vdume=V)

 

 

J
T

 

 

 

%

”it

TMOlZ Won

Figure 3.8 Electric and Magnetic Field Patterns for the Cavity Applicator

III

sir

EXC

thI

tur

511

EX

Th

EX

di

hi

43

the TM012 mode except the cavity is adjusted to one half the resonant
length, i.e., 7.2 cm. The discharge is formed at one end of the cavity
adjacent to the nozzle and its length is approximately a quarter wave
length long. Advantages of this configuration are that the discharge
can easily be brought close to or in contact with the nozzle and the
discharge has a smaller volume and hence, smaller surface which

minimizes radial energy losses.

3.4.3 APPLICATOR TUNING

An important feature of the cylindrical cavity applicator is its
ability to focus and match (little or no reflected power) the incident
microwave energy into the discharge zone. This is accomplished with
single mode excitation and "internal cavity" matching. Single mode
excitation allows the focusing and control of the microwave energy into
the discharge zone. The matching is labeled "internal cavity" since all
tuning adjustments take place inside the cavity.

This method of electromagnetic energy focusing and matching is
similar to that employed in recently developed microwave ion sources
except that it is used at higher discharge chamber pressures.6°'51'62
The differences with this application are associated with different
excited cavity modes and the discharge itself; i.e. differences in

discharge shape and location and the discharge properties due to the

higher operating pressures.

men
Inch
Ilscl
elec
curr

resi

the
varl
the:
are
its

liar

llltl
figl
CON

Vic-

44

The input impedance of a microwave cavity is given by

Pt + 2j0‘“. ' We)
2- I a R

+ 'X
l" ‘1 II I 2 J
2 O

(3.1)

 

in in

where Pt is the total input power coupled into the cavity (which
includes metal wall losses as well as the power delivered to the
discharge). WI and We are, respectively, the time averaged magnetic and
electric energy stored in the cavity fields and |I°| is the total input
current on the coupling probe. Rin and jxin are the cavity input
resistance and reactance and represent the complex load impedance as
seen by the feed transmission line. At least two independent
adjustments are required to match this load to a transmission line. One
adjustment must cancel the load reactance while the other must adjust
the load resistance to a value equal to the characteristic impedance of
the feed transmission system. In the cavity applicator the continuously
variable probe (8) and sliding short (2 of Figure 3.5) tuning provide
these two required variations, and together with single mode excitation
are able to cancel the discharge reactance and adjust the discharge
resistance to equal the characteristic impedance of the feed
transmission line.

This internal cavity matching technique can best be understood
with the aid of the equivalent circuit shown in Figures 3.9. These
figures display a standard circuit representation for a cavity which is
connected to a feed waveguide or transmission line and is excited in the

vicinity of a single mode resonance.63 Gc, Lc' and Cc represent the

45

Equivalent Circuit of the
Microwave Cavity Applicator

NM 1"
i viii Ii
Zin
i _Lc cc Gc pL C§L

OSCILLATOR FEED CAVITY APPUCATOR CIRCUIT DISCHARGE

TRANSMISSION IMPEDANCE
UNE

__ i n_ __ mg veff mg
g _ 5 +16 — EOE (“Eff + ((12):! +j€° [:w (vsz + 1132):]

Losses in Single Mode Cavity

G(:c1ciitﬂ.lHl2ds=Pb GL0: (If f fe"|E|2dv_-_ pa
5 v._
Stored Energy in the Field

2
con: -%ff€olE|2dV L00: {.fffmlnl dV

v-vL v-vL

, 2
)3L a: iffﬁ |E|2dv+ulyffﬂtlw| dV
v._ vL

Figure 3.9 Equivalent Circuit of the Cavity Applicator

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

46

conductance,inductance and capacitance respectively of the excited made
near resonance and jx represents the reactive effect of the evanescent
modes far from resonance. The relationships between the cavity fields
and these equivalent lumped circuit elements is shown in Figure 3.9. In
a cavity without a discharge, e'seo, and VL =0 and integrations for cc,
and Lc are over the entire cavity volume V. At resonance, the
capacitive and inductive susceptance cancel, resulting in a pure
conductive input admittance. The coupling probe (or aperture) is
represented in Figure 3.9 as the ideal transformer of turns ratio m:1.
Both circuit elements and the transformer are drawn with arrows to
indicate their variability during the tuning process.

The discharge is ignited by first adjusting the probe and cavity
length positions to excite a specific empty cavity resonance (i.e.,
TM012 or TM011) and to match the empty cavity applicator to the input
transmission system. Microwave power is then applied, absorbed into
the cavity without reflection and a discharge is ignited even with low
input powers of 20-50 W if the pressure in the discharge zone is reduced
to 0.5 to 10 Torr. The presence of the discharge changes Lc' Gc' and Cc
and adds an additional discharge conductance GL and susceptance, jBL, to
the circuit. That is, in the presence of a discharge a" and VL are no
longer zero and hence jBL and GL are also not zero.

As indicated by the equations in Figure 3.9, these equivalent
circuit elements are nonlinear functions of many experimental variables.
These include discharge gas mix and type, pressure and flow rate,
discharge geometry, absorbed microwave power (i.e.,lElz) and discharge
properties such as electron density, and collision frequency. The

nonlinear behavior of the discharge (and hence the behavior of the

47

equivalent circuit elements) is exhibited as hysteresis64 in
experimental variables such as input power, tuning, and operating
pressure.

The discharge admittance shifts the resonance, unmatching the
plasma loaded cavity from the feed transmission line. If the cavity
length and coupling probe remain fixed, further increases in incident
power result in only a slight increase in absorbed power and a small
change in discharge admittance since the cavity is further detuned from
resonance. Thus, the presence of the discharge allows only a small
portion of additional incident power into the cavity causing a large
increase in reflected power. This limited variation in discharge
properties is a fundamental problem associated with sustaining microwave
discharges in fixed size and fixed coupling cavities.54 Discharges in
these cavities can be only maintained over a very narrow range of
discharge loads (discharge densities, volumes, pressures, flow rates,
etc.) and thus these cavity applicators often operate with large
reflected powers.

The variable "internal cavity" matching employed in this
applicator provides the variable impedance transformation that allows
the discharge to be matched over a wide range of discharge loads. The
tuning together with variation of the incident microwave power "pulls"
the discharge properties along a discharge "loss line" similar to that
described elsewhere for cylindrical discharges.48'54 For a given
incident power, gas type, gas flow rate, and discharge pressure, i.e.,
for a given operating condition, the length and probe tuning are varied
in iterations until reflected power is reduced to zero. Typical tuning

distances are of the order of several millimeters and thus the tuning

48

process can be quickly performed either manually or with small motors
and can also be utilized as a simple discharge power control technique.

The matching is accomplished without altering either the plasma
shape and position or the mode electromagnetic field patterns and
without losing microwave power in external (conventional) tuning stubs.
Increases in input power increase the electric and magnetic field
strengths. However, the geometry of field patterns as shown in Figure
3.8, i.e. the electromagnetic focus, remains approximately constant
throughout the tuning process keeping the location of the plasma in the
center of the cavity away from external walls. Thus, the cavity system
can be tuned to a match as the experimental conditions, such as flow
rate, pressure, discharge configuration etc., change.

This tuning provides another important practical function.
Certain stabilizing gas flows require the insertion of a quartz center
body into the discharge chamber and into the cavity excitation region.
The position of the body was varied to optimize performance. Its
presence and position inside the cavity would tend to detune the cavity.
However, only a slight length tuning is required to compensate for its

presence and to return the system to a power match.

3.5 QICROWAVE ARCS

This section will describe the discharge itself and the mechanism
by which it is sustained and transfers energy to the propellant. First
the energy paths that convert microwave energy into thrust are
presented. The descriptions in these sections are brief and very

general. For further information on microwave discharges the reader

should refer to MacDonald55 and Marec66 and for information on general

in
'h

III

49

arc properties the reader should refer to Pfender57. Actual
measurements of the energy distribution of a microwave discharge at
pressures below 10 Torr, sustained by a cavity applicator and using
flowing hydrogen gas, are presented in Chapmansa.

In the last section experiments with these discharges are
presented under the conditions necessary for an electrothermal thruster.
The coupling structures presented in the section serve to match the‘

microwave power into the microwave discharge.

3.5.1 DESIRED ENERGY PATHS

The energy paths of a microwave electrothermal thruster are shown
in Figure 3.10. Microwave energy is transferred to the electron gas via
electromagnetic force interactions between the electrons and the
electromagnetic field in the applicator and is described in
Cherrington69 and Marec66. Ideally all of the microane energy is
transferred to the electron gas but in a real applicator some of the
energy is always lost due to resistive heating of the applicator
structure. The rate at which the energy is transferred to the
applicator walls is given as Pb in Figure 3.9 and is equal to the
surface integral of the tangential magnetic field over the interior of
the cavity surface. Chapter 5 presents measurements of this loss.

Due to the large differences in mass, electrons are accelerated
much more than the ions and therefor energy transfer to the electrons is
much greater than the energy transfer to the ions. Thus the
electromagnetic energy is transferred almost entirely to the electron
gas and the electromagnetic interactions with the ions can usually

(except at special ion resonances) be ignored.

50

Energy Paths

Desired Energy Paths
Converting Microwave

 

EnergyI into Thrust

 

 

Microwave Energy

 

 

I

I

 

Electron Gas

 

 

 

 

i

Undesired Energy Paths
that Result in Energy
Lost from the System

 

I

 

 

Applicator Walls

Resistance Heating Losses

 

 

Elastic Collisions
and
Inelastic Collisions

 

 

 

 

 

I

 

PropeHant
Thermalization

1. Elastic Collisions
2. Deexcitatlons
3. Racombinatlons

 

i
I
I
I
l
I
I
l
i
I
l
I
I
l
I
i
l
l
l
I
I
I
l
I
l
l

 

 

 

l

I

Losses due to
1. Radiation
2. Conduction
3. Convection

I

 

 

 

 

 

 

 

Nozzle

 

—+

 

 

Losses due to
1. Frozen Flow

2. Nozzle Imperfections

 

 

Thrust

 

 

 

Figure 3.10 Energy Paths Converting Microwave Energy into Thrust

 

 

51

If the direction of the electromagnetic field changes every half
cycle, the direction of force on the electron also changes every half
cycle causing the electron to oscillate with the electric field. If the
electron does not collide with a heavier neutral or ion or the walls of
the discharge chamber, it just oscillates out of phase with the field
and there is zero net energy transfer to the electrons over a complete
cycle. This means that no net energy is supplied to a particle in a
steady state time harmonic electromagnetic field when there are no
collision processes.

However, when there are elastic collisions with heavier particles
present the transfer of electromagnetic energy to the electron gas is
enhanced. Thus in a plasma where neutral and ion gases interact with
the electron gas, the electrons on the average gain energy after each
elastic collision and hence a net transfer of energy from the
electromagnetic field to the electron gas takes place. The time average
power density delivered to the plasma from the electromagnetic field is
given by

2
v ffNo(r)e
<p> =1 e IEI2 (3.2)

2 me(“eff2+wz)

 

where No(r)= electron density
an excitation frequency

veffa effective electron - neutral
collision frequency

(El: magnitude of the electric field
as electron charge
or per average electron
u e2
1 eff

2
<pe> =_ is] (3.3)
2 me(”eff2+“2)

 

52

Thus the time average power delivered to the discharge is given by the
equation for Pa from Figure 3.9.

The heated electron gas transfers its energy to the heavier ion
and neutral gases via elastic and inelastic collisions. Each typical
elastic collision, similar to a collision between two elastic balls of
unequal mass, transfers a small amount of energy (proportional to me/M
per collision) from the higher energy electrons to the heavier, cooler
ion and neutral gases. This energy is converted quickly into thermal
energy of the heavier gases after inter-ion and inter-molecular
collisions.

Inelastic collision also transfer energy from the electron gas to
the heavier gases by such inelastic processes as excitation, ionization
and dissociation. This energy is stored in the heavier gases until
additional inelastic collisions occur. Some of these inverse inelastic
processes are collisional recombination and deexcitation. These
deexcitation and recombination collisions release the ionization,
dissociation and excitation energy to the interacting particles. By
further collisional processes this energy is converted into the thermal
energy of the propellant.

Thermalization of the propellant results directly from elastic
collisions and indirectly from inelastic collisions through deexcitation
and recombination of ionized and dissociated species. A converging-
diverging nozzle then serves to convert the thermalized energy to thrust
energy. Ideally all of the electron gas energy is converted to thermal

energy and is available for the nozzle to convert into thrust energy.

53

In a real thruster, radiation, convection and conduction prevent all of
the electron gas energy from being available for conversion by the
nozzle.

Addition energy losses occur in the conversion of the thermalized
propellant in to thrust. Frozen flow losses occur when excited,
dissociated and ionized species flow out the nozzle without being
thermalized. Non ideal nozzle conversion of thermal energy into thrust

energy results in further energy losses.

3.5.2 DISCHARGE PROPERTIES

The particle and energy flow with respect to the surrounding
boundaries are shown in Figure 3.11. For this engine to operate cold
input propellant is fed into the discharge chamber. Microwave energy is
provided via the energy applicator. The microwave discharge serves to
couple this energy into the thermalization of the propellant. The
previous section discussed how the microwave energy in the
electromagnetic field is converted into thrust. This section discusses
properties of the discharge itself and where the energy transfer is
occurring.

The transfer of microwave energy into the electron gas occurs
within the discharge zone via the energy paths discussed in the last
section. This electromagnetic energy transfer and resulting energy
paths inside the discharge zone result in a large amount of
thermalization within the discharge zone and downstream from this zone.
At high pressures when the discharge is separated from the discharge
chamber walls, the discharge is at a considerably higher temperature

than the gas outside the discharge zone. Thus the microwave discharge,

54

Microwave Discharge Properties

Microwave Applicator —l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

____I_J 7 Cold f
.... Propellant
Microwave —" /
Energy _"‘" Radiation
_' Microwave
: Discharge
«‘
Outward Flow of
Charged. Excited
and :
Heated Species ..
Radial Heat
Convection and
Conduction
Inward Flow of
Cold Neutral and l
Recombined
S i .
pic °° Th erm al Ized 02:33};
Propellant and Wall
Stabilization
J

 

 

 

 

 

 

 

ll .

Discharge
Chamber
Nozzle
(Must;

I

Figure 3.11 Microwave Discharge Properties

 

55

like low frequency arcs, is a thermally inhomogeneous discharge with a
hot central core and sharp temperature gradients between the center and
the surrounding cooler gas outside the discharge region. If the
pressure is assumed to be uniform across the discharge chamber, the
density in the discharge must be lower than the density outside the
discharge because of the high temperature in the discharge zone.

As shown in Figure 3.11, there is a flow of cold neutral and
recombined species into the discharge zone. At the same time there is
net outward flow of ions, electrons, dissociated atom and thermally
excited atoms. Under steady state conditions the density remains
constant so that these two flows must be equal. In addition, because of
temperature gradients, there will be convection flow from the hot center
to the cool surrounding gas and heat conduction to the surrounding gas.
Further if there is gravitational forces present, convection currents
can occur because of buoyancy effect of the lighter heated gas verses
the heavier cold gas. These flows are important because they bring cold
neutral atoms into the discharge to be heated and they transport energy
out of the discharge heating the propellant. At the same time these
flows, along with radiation from the discharge, present a energy loss to
the boundary walls.

At low pressure (510-100 Torr) the ions, electrons and
disassociated atom will diffuse to the wall of the discharge chamber
where wall recombination takes place. At low pressures the particle
density is not large enough for volume recombination to occur at a high
rate. As the pressure is increased the rate of volume recombination
increases. The increase in recombination together with the large

thermal gradients causes the discharge to constrict.7o This

56

constriction is very important in the presence of a flowing propellant
since the farther the discharge is away from the wall, the lower the
energy losses to the walls and the greater the transfer to the
propellant as it passes around and through the discharge.

In the applicators described in this thesis, the constricted
discharges are held in place regardless of the rate of propellant flow
by wall stabilization and the proper exciting electromagnetic field
(mode in a cavity applicator) configuration. The maximum energy
transfer to the discharge occurs with the discharge centered in the
chamber by design of the applicator. If the discharge moves off center
the energy into the discharge is reduced and the applicator is detuned.
At the same time the thermal energy loss to the chamber walls is
increased. Thus any movement off center reduces the energy of the
discharge, this reduction increases the particle density in the
direction of movement which intern forces the discharge back to the
center. This stabilization allows the discharge to be maintained off

the chamber wall again reducing the energy loss.

3.5.3 DISCHARGE INITIATION

The description above describes the discharge operation in steady
state condition. The methods used to start the discharge are also
important. Two methods, and combinations of them, were use in these
experiments. In the first method, the pressure of the discharge chamber
was reduced to the point where the effective collision frequency, “eff!
of the free electrons in the neutral gas, was equal to the radiation
frequency, w, of the microwave field. The applicator was then tuned for

maximum E field and a discharge was initiated by tuning for minimum

57

reflection. The pressure for this was between 0.5 Torr and 10 Torr for
most gases.

If a high enough electric field could be generated, a second
method was used at higher pressures. The applicator was tuned to
maximum B field and a Tesla coil was used to inject extra free electrons

in to the discharge chamber and breakdown occurs.

3.5.4 DISCHARGE EXPERIMENTS

This section describes research into the generation of high
pressure microwave discharges and the properties of these discharges.
The applicators described in this section produced discharges in several
gases at high pressures. Experiments with other gases showed the
ability to generate discharges in them, but were not studied at high
pressure.

Experiments with hydrogen microwave discharges at atmospheric
pressures were conducted by Kapitza.71'72 These experiments used a
cylindrical resonant cavity 20 cm in diameter operated in a TMOln mode.
A variable power supply could provide power levels of up to 174 kw at
1.55 GHz although the maximum absorbed power was 20 kw. Filaments of
hydrogen about 1 mm in diameter and up to 10 cm long (at full power)
were formed inside the cavity. Measurement of electron density and
temperature were taken. Dymshits73 calculated a gas temperature of
9000 K from this data for the hydrogen discharge.

Experiments with both hydrogen and nitrogen were conducted by
Arata.”'75 These experiments used a rectangular resonator with a
variable power microwave source that was capable of power levels of up

to 30 kw at .915 Ghz. Power absorption was reported to be greater than

58

802 into the discharge. For a input power of 20 kW, nitrogen was
reported to have a gas temperature of 6300 K with a 4 cm discharge
chamber and a gas temperature of 6800 K with a 2 cm chamber. Hydrogen,
at a input power level of 20 kW, had a gas temperature of 9000 K.

Experiments with argon were conducted by Rogers76

using the
cylindrical resonator described in section 3.4 of this chapter. In a
typical experiment input power was 500 Watts at 2.45 GHz and a 12 mm
diameter quartz tube, running through the center of the cavity, was used
as a discharge chamber. For a typical experiment a 0.2 mm filament
discharge was produced with a power density of 130 W/cm3, The inferred
gas temperature was given as 1400 K with a collision frequency of
1.2x101o sec".

Experiments with argon were conducted by Hubert77 in a surface
wave coupler. For input power of 100 Watts at 1.7 Ghz and a 5 mm
diameter quartz discharge chamber, a 14 cm long discharge was produced
with a diameter of 1 mm. This discharge was at one atmosphere of
pressure and gas temperature was given as 2500 K.

Experiments with nitrogen were conducted by Miyake58 in a
microwave torch. A gas temperature of 6700 K at high pressure was given
for a input power of 2 kW at 2.45 GHz. At these conditions the
discharge was 10 cm long and about 0.5 cm in diameter. Further
experiments with a torch were conducted by Batenin.78'79 For a
atmospheric discharge in hydrogen, with a input power of 2 kW at
2.45 GHz, a gas temperature of 8500 K was measured. For a atmospheric
discharge in helium, with a input power level of 900 Watts at 2.45 GHz,
a gas temperature of 65003500 K was measured.

These experiments show that high pressure discharges can be

59

generated and sustained by microwaves. They also show that these
discharges are very hot and thus provide great potential for
electrothermal use.

There is some interest in using more complex materials as
propellants. These materials are left over from other spacecraft
processes and would otherwise go to waste. Bandel80 was able to
generate a microwave discharge in a mixture of air and water. Mertz81
used a microwave discharge of a mixture of CO and Hz to produce methane.
While these experiments were at low pressure, they demonstrate that

discharges can be produced in complex molecular gases.

CHAPTER IV

EXPERIMENTAL SYSTEMS AND MEASUREMENTS

4.1 INTRODUCTION

This chapter describes the experimental systems that were used to
generate and measure the data presented in this thesis. The first
section presents a general overview of the entire system. The second
section describes the experimental procedure and calculations used to
take measurements of properties of the high pressure discharges. The
third section describes the experimental procedures, measurements and
calculations used to evaluate the properties of microwave electrothermal

thrusters.

4.2 GENERAL EXPERIMENTAL SYSTEM

In order to experimentally create and control the microwave
discharge and make measurements of thruster properties, several
experimental measurement and control systems were necessary. The first
measured and controlled the flow rate and pressure of the test gases
used in the experiments. The second system measured and controlled the

microwave energy and the third system measured the plasma volume size.

4.2.1 GAS FLOW, PRESSURE AND VACUUM MEASUREMENT
A diagram of the basic gas flow system is shown in Figure 4.1. As

shown in this figure the system could be divided into two parts. The

60

61

Gas Measurement System

 

Flow Control

 

 

EL

 

 

 

 

 

Ls

 

L— Flow Transducers
Helium and Con trolers

 

 

 

 

 

 

 

 

Nitrogen Thermocouple
ermocouple Probe
Display {3"
Discharge
Chamber t.
h—
E Nozzle
Water In

 

Heat
Exchanger

 

Water Out :j

Flowing
Experiments

Thermocouple

 

ermocouple

b —_———

Probe

 

Display
Gate

 

 

 

Vacuum
Pump

 

{1—

Valve

 

H Manometer
._. Display
D-1D,DDD Torr

 

 

and Display

 

 

 

Electronic
—~ Manometer

 

 

 

  

 

 

 

 

 

 

 

 

U

Microwave
Applicator
__ll
Shutoff
‘ Valve
No-Fiow
Experiments
Manometer
~ ~ Display
0.1-1000 Torr
Electronic
Manometer

Figure 4.1 Gas Flow. Pressure and Vacuum Measurement System

 

“'st F
grits
'li till!
3525,

yessur

msun
m‘s in;
am
meril
“we
it‘iere
71,000
(*4 ch:
e‘ectrc
it gas

Fl
ﬁssure
midi
i elect
i’t'ssur
lit-met
‘0' con
.3 I935

amber

62

first part supplies the gas to the discharge chamber and the second part
serves to evacuate the discharge chamber. For all experiments presented
in this thesis the propellants used were helium, nitrogen and oxygen
gases, and were supplied to the system from high pressure cylinders with
pressure regulators attached on cylinders to reduce the pressure to
usable levels.

As shown in Figure 4.1 the system was capable of providing flow
measurement of the propellant if it was required. In early experiments
this was provided by a rotometer but was this found to be difficult to
control and calibrate and is not shown in the Figure 4.1. For latter
experiments a MKS 3 channel flow controller was used and is shown in
Figure 4.1. This controller could measure the mass flow of three
different gases into the discharge chamber, in a range of 50 to
10,000 sccm for each gas, and could also give the total gas flow into
the chamber in sccm. By use of a electronic feedback circuit and
electronic valves it could maintain a constant value of gas flow when
the gas pressure or other conditions in the discharge chamber changed.

For all experiments discharge pressure was a necessary
measurement. As shown in Figure 4.1 two different systems were used
depending on the experimental conditions. For conditions above 1 Torr,
a electronic MKS manometer and a Raise gauge were used to measure the
pressure in the discharge chamber, and a 0.1-1000 Torr Datametric
manometer was used to measure the pressure in the evacuation system.

For conditions below 1 Torr, a Hastings thermocouple type gauge was used
to measure the pressure in both the vacuum line and in the discharge
chamber.

The vacuum was provided by a two stage mechanical roughing pump as

iditl 1
atuUI !
tzzle,
uterinn
15 usec

gate: 1

L2.2 HI

ih
iittiOT
ntrouai
'iCWai
3* iIOi
‘9 S-ba
lite 0!

As
miller

in

‘ term
1536 by

63

shown in Figure 4.1 and a gate valve allowed the pump to shut off from
the discharge chamber evacuation line. In order to provide the maximum
rate of pumping, 2" vacuum Pyrex glass piping was used in the vacuum
system wherever possible. In experiments with no gas flow, the vacuum
system was used to evacuate the tube and a small amount of gas would be
allowed to flow through the system to flush out any impurities that
might be present. Then a valve is closed and gas is added to bring the
system to the desired experimental pressure. In flowing experiments the
vacuum system serves to keep the pressure low on the exhaust side of the
nozzle, and thus also serving to exhaust the propellant from the
experiment. In the experiments that required gas flow a heat exchanger
was used to cool the gas near the nozzle exit to protect the vacuum

system from the high temperature propellant.

4.2.2 MICROWAVE SYSTEM

The microwave system used in these experiments had two important
functions. The first of these was to deliver a controllable source of
microwave power to the microwave applicator and to safely handle this
microwave power. The second was to provide an accurate measurement of
the amount of power being coupled to the applicator and the discharge.
The S-band, rectangular waveguide microwave system that accomplished
these objectives is shown in Figure 4.2.

As shown in Figure 4.2, the system consisted of the following
components. A 2.45 GHz microwave source was used to provide a well
filtered continuous wave supply of microwave power. Different supplies
made by several manufacturers were used to provide different amounts of

microwave power needed in the experiments. The output of the microwave

64

Experimental Microwave
Supply System

 

 

Incident
P M t
Microwave ower e er Wasting/re
Source App c
, _ Attenuator
l in

[Power I I Directional J l Pi-—>
Divider [ [Coupler J C
_ Circuiator
Reﬂected
Power Meter

Matched
Load Attenuator
[k Directional l <——— 1’,
Coupler J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 4.2 Experimental Microwave Supply System

65

source was connected into a microwave power divider. This device
allowed a variable portion of the microwave signal to be absorbed into
an internal load. This proved necessary since some of the microwave
sources provided a stable, clean signal only near full power.

As shown in Figure 4.2 the output of the power divider was
attached to the input of a waveguide broadwall directional coupler.

This device couples a small amount of the input signal into the
secondary arm where a precision attenuator and a microwave power meter
are attached. By careful calibration with a source of known power and
frequency, the power reading on the power meter gives a very accurate
measurement of the incident microwave power.

Next the signal was inputted into the first port of a 3-port
waveguide microwave circulator. The second part in the circular
coupling structure was attached to the transmission system to the
microwave applicator. The transmission system to the applicator used
two different systems. For some early experiments a coaxial type system
was used, but for all later experiments a flexible waveguide system was
used because of less power loss and increased power handling capability.

The circulator provided at least 40 db of isolation between the
incident and reflected coupling signals. This isolation protected the
source from damage due to reflected signals and increased the accuracy
of the incident and reflected power measurements by eliminating the back
coupling into the directional couplers. The circulator also minimizes
frequency pulling of the magnetron by variations in the plasma load.

The third port of the circulator was attached to a second
broadwall type directional coupler. A precision attenuator and

microwave power meter again were used to measure reflected signal power.

66

The output of the directional coupler terminated in a water cooled dummy
load. This load served to absorb the microwave power not coupled in the
energy absorption chamber.

By inputting a microwave test signal of known power and measuring
both at the incident power meter and replacing the applicator with a
second power meter an accurate calibration of the input power Pi can be
made. By attaching a microwave source of known power in place of the
energy absorption chamber and measuring the power at both the reflected
power meter and a power meter attached in place of the water load, a
very accurate calibration of reflected power Pr was made. Thus the

power absorbed in the energy applicator is given by Pt'Pi'Pr°

4.2.3 DISCHARGE VOLUME MEASUREMENTS

The size of the plasma volume was a desired measurement in some
experiments. A modified technique of Rogers13 was used to make these
measurements. A diagram of the setup is shown in Figure 4.3. As shown
a camera is used to take a picture of the plasma discharge. In order to
reduce the amount of aberration distortion in the resulting picture,
Rogers technique was modified. Instead of a singlet close up lens, a
rectilinear wide angle lens with a extension tube was used to take the
picture. As shown in the figure this allowed the camera lens to be
placed against the window of the cavity applicator and yet because of
the wide angle lens the entire discharge was visible in the picture.
TYPically a 24 mm or 28 mm lens was used to take the picture and from
1 mm to 10 mm of extension tube was necessary depending on the focusing
ability of the lens. For all gases the same exposure was used, EV12

"1th 8 f/stop of 4 and a speed of 1/250 of a second or a f/stop of 2.8

67

Discharge Volume
Measurement Setup

Cavity DZ ‘:1
Applicator

24mm or

 

 

 

 

 

 

Plasma
Discharge

 

 

 

 

 

1
Viewing F
Window
‘—.

 

 

 

 

 

28mm Lens

J

35mm
Camera

 

 

 

 

 

1—10mm Extension Tube

Figure 4.3 Experimental Setup for Discharge Volume Measurement

.c'

68

at 1/500 of a second.

In order to calculate the area of the plasma, slide pictures were
taken using Ektachrome 200 film. With no discharge present a piece of
graph paper was inserted into the discharge chamber in approximately the
same position as the discharge. A picture of this graph paper was taken
and used to correct for spherical distortions due to the wide angle.
Then the plasma volume was calculated by first taking a picture of the
discharge with the camera set up as mentioned earlier. Next the slide
was projected onto a large piece of graph paper with the discharge
enlarged as much as possible. The discharge was then divided into equal
height solid disks and the diameter of each disk was measured. These
diameters were then corrected for the distortions mentioned earlier.
Then the volume of each disk was calculated by computing its volume as a
solid cylinder and the total plasma volume was computed by summing the

volume of all the disk sections.

4.3 algg PRESSURE CAVITY DISCHARGE CRITERIA

Experimental quantities of interest in the evaluation of
applicator coupling performance are the power absorbed in the discharge,
the microwave coupling efficiency to the discharge and the loaded
cavity 0. By limiting the discussion to a single mode in a cavity type
applicator, namely the TM012, and by using the technique of Rogers13,
the coupling efficiency, loaded cavity 0 and microwave power absorbed in
the cavity walls can be calculated from the empty cavity 0 and empty
cavity absorbed power. A brief review of this technique is presented in

the next sections .

69

4.3.1 COUPLING EFFICIENCY

The difference between the incident power, Pi, and the reflected
power, Pr' measures the total power, Pt, delivered to the applicator.
Power delivered into the applicator divides itself between the power
delivered to the conducting cavity walls, Pb' and the power delivered to
the discharge, Pa° Thus Ptan+Pa. These two quantities can be related
to the excited single mode cavity fields, discharge variables such as
plasma frequency, mp, and the effective collision frequency, ”eff' and.
the intrinsic resistance of cavity walls. The exact division of the
power between the walls and the discharge depends on the relative
lossyness of the discharge vs. the lossyness of the cavity walls.

It is useful to define a system "figure of merit" called coupling
efficiency, which is concerned with the efficiency of coupling microwave

power into the discharge. The overall coupling efficiency can be

defined as

p

 

(Eff)1 . x 100 (4.1)

Pi

where Pi'PrIPt'PrTPaTPb' Viewing the applicator as an impedance
transformer and focusing device, an ideal applicator will deliver all
the incident power into the discharge with zero reflected power and
applicator wall power loss. Thus the overall coupling efficiency will
then be 1008. In most experiments the reflected power can be reduced to
a very small amount by tuning adjustments, so that Pr“Pi' Then the

overall coupling efficiency is equal to just the applicator coupling

70
efficiency
Pa P

a
8100 x —— (4-2)
Pt PaTPb

 

(Efflz = 100 x

Despite the simplicity of this equation the coupling efficiency is a
difficult quantity to determine experimentally since the wall losses are

difficult to measure.

4.3.2 CAVITY LOADED Q

If the cavity applicator is assumed to operate in a single mode a
general equation can be derived for the power absorbed in the plasma.
The microcoax probes mentioned in a earlier section were used to sample
the radial electric field at the wall of the cavity applicator. The
small size of the microcoax probes allowed them to be inserted and
removed during actual experiments with little detectable perturbation to
the plasma or cavity fields. The measured loaded cavity axial field
distribution agreed well with the theoretical TM012 mode distribution of
'Erl and is shown in Figure 4.4.

It is assumed that the presence of the discharge does not
significantly alter the spatial distribution of the cavity wall currents
from those of the empty cavity TM012 mode. Evidence supporting this
assumption is that only very small experimental changes in resonant
length are required to match the plasma from the empty mode to the mode
with the discharge present as shown in Figure 4.4. The field
distribution with the plasma present varies only slightly when compared
to the empty cavity as measured by the microcoax probes. The exact

numerical solution for the cavity field distribution with the lossy

7]

Radial Electric Field Versus
Position for a Nitrogen Discharge
25mm i.d. Discharge Chamber Oriented Vertically

 

Measured Power - mW

 

L5 - 1.8 cm input Power - 370 Watts
LP - 13.93 cm Chamber Pressure - 305 Torr
. ............ _ Calculated
4 -i O — Measured
- ...; 0 ...: ‘0' ...:
5' °'-. 0' '°-.
..0
2 ‘ <3? 0
“ b
1 - o
- O
O 4.. I i I [ G i T r I

 

 

0.0 0.2 0.4 0.6 0.8 1.0
Position Relative to Sliding Short

Figure 4.4 Radial Electric Field for a Nitrogen Discharge

72

plasma present only differs from the empty cavity distribution near the
discharge in the center of the cavity13. Under these conditions, the
ratio of the radial electric field measured at a fixed position on the

cavity wall to the total power absorbed by the wall is a constant with

and without a discharge. Thus

Pb

 

a constant (4.3)

iErl

By measuring the power absorbed Ptov the cavity quality factor,
Quo' and the associated radial electric field era for the critically
coupled empty cavity excited by a low power test signal, the absorbed

power in the wall, Pb' and Qu for a cavity loaded with a discharge can

be determined from the following equations

15.42
Pb . ———————— Pto (4.4)

2
lawn
and

2
0.. Pt. lerl

Qu 3
IErol2 Pt

(4.5)

where Er and Pt are respectively the radial electric field and the
cavity absorbed power with the discharge present. Since both of the
equation require the ratio of electric fields, only the relative
magnitudes of the electric fields are necessary. Thus the electric

field probes do not have to be calibrated for these measurements.

4.4 THRUSTER PERFORMANCE CRITERIA
The performance of a thruster is measured in terms of thrust,

Specific impulse, and energy efficiency. The equations used to derive

73

these quantities are presented in Chapter 2. All of these equations,
2.1-2.5, are derived from the measured thrust, therefore the method used
to measure the thrust is of great importance since all other
calculations will depend on it. The best method of thrust measurement
would be direct thrust measurement using the vacuum of space under
weightlessness conditions. However the availability of space shuttles
and nonreusable boosters, and the their cost of operation limit almost
‘all thrust measurements to ground experiments.

The experimental techniques for thrust measurement used in this
thesis were developed by s. Nakanishi of NASA Lewis Research Center and
described here. The next sections present two different ways to measure
thrust. The direct thrust measurement system used a thrust stand to
take direct thrust readings. The indirect thrust measurement system was

used to take thrust readings when a thrust stand was not available.

4.4.1 DIRECT THRUST MEASUREMENT

For direct thrust measurement a high resolution, high sensitivity
thrust stand was used. This system was set up in a vacuum tank at NASA
Lewis Research Center so that a direct measurement of thrust could be
taken with a test engine exhausting into a high volume tank and pumping
system.

This system used a precision thrust stand shown in a front view in
Figure 4.5. Figure 4.6 shows a side view and both views show the
relationship of the center of the thrust target to the center of the 4
inch test port of the vacuum tank. The thrust stand is a torsional
pendulum in which the target is balanced by the torsion in the

suspension wire. A counterweighted arm with the target on one end and a

74

Side View of Thrust Stand

 

3
—- Vacuum Tank 7
r—Torsion Head

KM] -—— Gear Mater

L— Music Wire

 

 

 

r- Thrust Target .——Gulde Tube

Calibration Rig

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. l--l
4 Vacuum
Flange
\’ — Calibration Weights
Monoﬂioment Thread —-
Center Line
for 4" Vacuum

 

 

Flange and
Thrust Target |
Pre-Laad Torque Head

 

\
5

7

Figure 4.6 Side View of Thrust Stand

75

Front View
of Thrust Stand

{—— Torsian Head
E /— Gear Mater

-— Music Wire
~— Cuide Tube Thrust Target
Magnetic Damper Outline of 4"

J— Counter Weight Vacuum H0090
Opening

 

   

 

 

\\

 

 

 

_,_ ﬂ“
age.

 

 

—- Damper Plate
-Opticai Position Sensor

 

 

 

./ ll=

 

 

\

 

 

 

 

 

 

 

 

i
E Vacuum Tank Wall

i_ \

Pre—Laad Torque Head

 

 

Figure 4.5 Front View of Thrust Stand

76

magnetic damper plate on the other end is suspended by a single strand
of music wire clamped at the middle and selected according to the range
of thrust to be measured. Initially a 0.76 mm in diameter wire was
used, but due to an increase in the range of thrust measurement that was
required, the wire size was increased to 1.04 mm to accommodate these
higher thrust levels. Application of a counter torque by the upper
torsional head driven by a gear motor restores neutral equilibrium to a
reference position monitored by a optical position sensor sighting the
damper plate. The reference position corresponds to a condition where
the thrust vector is perpendicular to the target plate. The thrust can
be measured as a function of the twist in the upper wire required to
maintain neutral equilibrium under any thrust condition on the target.
Frictionless calibration of the thrust measurement system was
accomplished by a rig which established an equilibrium of forces such
that the horizontal force equals the vertical weight transmitted via a
thin monafilament thread. A windlass permitted the application and
removal of calibration weights to the back of the target. The wire
twist was resolved by a 10-turn potentiometer attached to the worm shaft
of a 20:1 reduction worm gear drive. Thrust measurement were always
made with the worm driving in the same direction to eliminate back lash
errors. The wire twist measurement corresponded to a 11.65 degree per
Volt to permit a resolution of 0.01165 degrees when read on a 1 mV range
on a digital voltmeter. The sensitivity of the optical position sensor
was 0.47 V per degree of wire rotation. The target could be routinely
adjusted to reference position within 15 mV or 0.0106 degrees. Since
the thrust calibration constant with a 1.04 mm wire was 0.018 Newton per

Volt on the worm shaft, or 0.018 N for 11.65 degrees of wire twist, the

77

thPUSt OQUIVSIODt 0f 1-545X10‘5 N per degree of wire twist implies that

a reference position error oi"__+_1.61‘r10‘5 N.

4.4.2 POSSIBLE ERRORS IN DIRECT THRUST MEASUREMENTS

There are several possible sources of error with this thrust
stand. The first possible source of error is due to the center of
thrust of the test thruster not being properly aligned with the thrust
target. To limit this problem the center of the thrust stand target was
carefully aligned with the center of the vacuum port. Further, the
target plate was carefully adjusted so that it was parallel with the
mounting face of the vacuum port. Finally to test if the thruster was
off center in its thrust axis, it was rotated through 90 degree turns
and operated with a cold gas flow to test for variations in thrust. No
variation could be measured, so it was assumed that the thruster was
operating with its center of thrust axis properly aligned with the
target plate.

The next possible source of error is from scattering between the
exhausted propellant of the thruster and gas molecules in the vacuum
tank. During thruster tests, it was impossible to maintain ideal vacuum
facility conditions for thrust measurement. For some tests, direct
thrust measurement was not obtainable at all. A short series tests were
made to evaluate the affects of vacuum tank pressure on thrust
measurement and to develop a method for calculating performance even
without measurement of thrust. These tests consisted of thrust
measurements made at a constant propellant flow rate and temperature
while the facility pressure, as measured by an ionization gauge, was

varied by bleeding in nitrogen gas at various rates. A set of thrust

78

measurements were also made at a constant propellant flow rate and
microwave power, while the facility pressure was varied.

Measured thrust as a function of facility pressure is shown in
Figure 4.7 for several operating conditions. All curves except one are
for isothermal flow of nitrogen gas with a temperature of 296 K. From
the theoretical specific impulse equation, nitrogen at this temperature
should yield an Isp of 80 seconds. Using this Isp' the theoretical
thrust corresponding to each propellant flow rate has been calculated
and they are shown as dashed lines in Figure 4.7. At the lowest flow
rate of 29x10'6 kg/s, the lowest no bleed facility pressure of
6.3x10'4 Torr was obtained. The measured thrust was within 10'3 N of
the theoretical thrust. As the facility pressure was increased with
nitrogen bleed, the measured thrust decreased as much as 338 for a
decade increase in pressure. Similar trends were observed at other flow
rates and also for the case in which 294 Watts of microwave power was
absorbed in the plasma discharge.

In the isothermal flow cases, a extrapolation of the curves allows
the intersection of each with the theoretical thrust value line
corresponding to that flow rate. It is observed that these
intersections occur at pressures near 5x10‘4 Torr. This implies that if
facility pressure could be maintained below 5x10‘4 Torr at all times,
nozzles of the type used can be expected to achieve almost 100%
efficiency. The curve shown for 294 Watts of microwave power by similar
extrapolation produced a a thrust of 6.6x10'2 N to yield an ISp of

113.4 seconds.

Thrust -4- our

79

Th ru st Versus

Background Tank Pressure

 

 

 

 

T H t -
\<><>O 54x10"6 kg/s
5" 72x10"6 kg/s Cold \ -6
O
C Id
— .59110'339A £06., _ _€°\ \Q)~
4— 48x10"6 kg/s Cold \%\ \A_4
____‘<E—_J%ﬁ5
.... D\E} _
‘Ci
2e _2§ﬂ°'26_k9_/2. H POE ... _.2
o=o\o
\O
-i T -
O lillrilii illiilllo
10“ 10"3 10‘2

Background Tank Pressure — Torr

Figure 4.7 Thrust Versus Background Tank Pressure

Thrust - Newtons x10”2

80

4.4.3 INDIRECT THRUST MEASUREMENT

In some experiments direct thrust measurements are not available.
For instance, in some experiments a high volume vacuum tank was not
available, the nozzle could not be positioned to operate the thrust
stand correctly, or the thrust stand did not have enough range. At
these and other instances when direct thrust measurements are not
available, an indirect means of thrust measurement was used. This
section presents this indirect method of thrust measurement.

If it is assumed that a nozzle of fixed throat area is always
choked at a given propellant flow rate, the upstream pressure varies as
the square root of the gas temperature. Because specific impulse also
varies as the square root of the gas temperature, the thrust at a given
propellant flow rate varies directly as the upstream pressure. The
assumptions inherent in the above reasoning is that the gas is in
equilibrium and that the nozzle efflux is uniform and one dimensional.

Thus if the thrust is directly proportional to upstream pressure
for a given constant flow'rate, a linear relationship can be assumed
between the hot to cold thrust ratio and the hot to cold pressure ratio.
If a direct proportionality is assumed conservatively, the thrust and
specific impulse of any gas can be calculated. The theoretical cold

specific impulse can be calculated from the equations

Ispc a 24.6(To/M)‘/2 (4.6)
for diatomic gases and

Ispc = 20.9(T0/M)1/2 (4.7)

for monatomic gases. Where To is the gas temperature prior to expansion

and M is the molecular weight of the gas. A full isentropic expansion

81

to vacuum conditions is assumed.
The hot specific impulse can be obtained from the measured hot to
cold pressure ratio, because for a given propellant flow rate, the

specific impulse is proportional to thrust. Thus,

 

IspH pH
. .— (4.8)
Ispc pc

Knowing the hot and cold specific impulse, and the thrust power, the

efficiency can be calculated from equation 2.4 as,

lZZlZ

n a (4.9)
zpa

 

 

1

 

+
gziirspciz

and the power to thrust ratio is given by substitution of this value of

efficiency in equation 2.5.

4.4.4 VERIFICATION 0F INDIRECT THRUST MEASUREMENTS

It is clear that because of pumping limits, facility pressures
below 5x10“ Torr will seldom be achieved, especially at high propellant
flow rates. At these and other instances when direct thrust
measurements are not available, the indirect thrust measurement
technique is used. The applicability of the indirect thrust measurement
technique for calculating thrust was examined by plotting the ratios of
directly measured thrust and the tube pressures. The "hot" and "cold"
conditions are with and without microwave power, respectively. The

ratio of hot to cold measured thrust is shown plotted against the hot

82

and cold pressure ratio in Figure 4.8. Data from both the cavity
thruster and the coaxial thruster configuration (presented in latter
chapters) are shown for a wide range of facility pressures. There is a
linear relationship between the two hot to cold ratios. There is a
tendency for the thrust ratio to be about 38 higher than a direct 1:1
proportionality with pressure ratio. It is not known whether or not the

thrust measurement tended to be 38 high during hot conditions.

83

Comparison of Hot to Cold Pressure
Ratio Versus Hot to Cold Thrust Ratio

FHot
F Cold

Measured Thrust Ratio —

 

 

 

 

 

Tank
Point Configuration Pressure
Torr
O Coaxial 0.9x1 0"3
CI Coaxial Liana"
A Coaxial 1 . 4x1 0-3
<) Cavity 4».8x10"'3
o Cavity 3.0xio"
/
/
4. ._ /
/
/
/
/
/
/
3 /
_ 663’
/
/
of
1 T /
/
/
/
/
O l i l l
. F)Hot
Chamber Pressure Ratio -
PCold

Figure 4.8 Comparison of Pressure Ratio to Thrust Ratio

 

 

 

CHAPTER V
HIGH PRESSURE MICROWAVE DISCHARGES

5.1 INTRODUCTION

The results of experiments with high pressure microwave discharges
are presented in this chapter. These experiments were conducted for
several reasons. First to experimentally produce high pressure
microwave discharges in different gases and measure the microwave
coupling efficiency to them. Second, to determine the tuning conditions
necessary to produce and maintain these plasmas. Third to measure
discharge properties vs. pressure for different diameter discharge
chambers.

The experimental microwave energy coupling and measurement system
variations from chapter 4 are first described along with a description
of how the experiment was conducted. Measured properties such as
pressure, power density, coupling efficiency and cavity 0 are presented

for variation in pressure and chamber diameter.

5.2 EXPERIMENTAL SYSTEM

The experiments were conducted using the general techniques and
equipment described in chapter 4. However the specific setup and
techniques used to conduct the experiments and measure the data is
presented in the next two sections. The first section describes the
microwave setup and use of the microwave applicator. The second section

describes the general experimental setup and procedure.

84

85
5 2.1 MICROWAVE CAVITY TYPE ABSORPTION CHAMBER
The caVity type absorption chamber is shown in Figure 3. 5, and its
° The cavity applicator was mounted in
ning

Fused quartz tubes run

own in Figure 3.5.
r served as discharge chamb

probe actuato

operati
ers for the

anner as sh
ribed in

a vertical m
applicato

'ally through the
r designs desc

coaXi
The advanc

re not employ

ed short and
A variable power

ed in these experiments.

experiments.
1, we
45 CH: m

section 3.4.
icrowave sour

500 Watt CW 2.
power for these experiments.

2 was conduc

averaged
The empty cavity measurement of \Erol
power microwave source to prevent breakdown in the caVity The reading
were averaged in the same manner as the earlier reading of \Eri2 The
empty cavity measurement of Quo 5 taken using a sweep generator and
measuring bandwidth between the half power points on the applicator
resonance curve.
ROCEDURE

5.2.
The basic vacuum and gas hand
experiments were described in chapter 4. Cylindrical fused quartz tubes
as shown in Figure 3 5, and were

86
designed to be interchangeable. For these experiments three different
diameters were used. They were inside diameters of 12 mm, 25 mm, and 37
mm. The method of evacuation mentioned in chapter 4 was used to
evacuate the discharge chambers. Gas was slowly added to the discharge
chamber to reach the desired pressure level. A slight flow was added
and the microwave power was turned on. A Tesla coil was used to start
the discharge in the applicator. The flow was turned off and the
pressure raised to the desired level. Until the pressure was above 40
Torr, the cavity tuning was set for 40 Torr to prevent the discharge
from 90109 OUt- The values 0* Ls and Lp for this are shown in Tables
5.1-5.3. Above 40 Torr the tuning was adjusted for maximum power
absorbed in the applicator. Compressed air blowing on the discharge
chamber from a second part was used to cool the chamber. The camera for
the measurement of discharge area was attached to the cavity for
stability and set up to measure plasma volume using the technique of
section 4.2.3. Figure 5.1 shows photographs of a nitrogen discharge as

the pressure is varied.

5.3 EXPERIMENTALtggASUR§M§31§

For these experiments the following measurements were taken. The
discharge chamber pressure, microwave incident and reflected power,
lEr|2, Lp (input probe length), Ls (input short length), and discharge
size and are presented in Tables 5.1-5.3. Three different gases were
tested, oxygen, nitrogen and helium. Using the equation 4.2 for
coupling efficiency and the technique described in section 4.3, coupling
efficiency was calculated for nitrogen and helium in each of the three

tubes. Figure 5.2 shows the results for nitrogen. A coupling

 

87

Discharge Constriction
Versus Pressure

Nitrogen Gas in a 12mm i.d. Cylindrical
Quartz Discharge Chamber

Pressure —- Torr

4O 72 108 148 196 760

Figure 5.1 Picture of Discharge Constriction Versus Pressure

40

72

‘ 108
148
196
257
340
474
584
760

40

72

108
148
196
257
340
474
584
760

immmma

40
72
108
148
196
257

88

Table 5.1 Experimental Results in 12mm Discharge Chambers

Nitrogen in 12mm Tube, 1/29/1983

pt (5', (2 pb 04 (Eff)2 Vogume (p) 3 ts Lp
Watts (Watts) Watts - 8 cm W/cm cm on
468 8.0 10.6 100.4 97.73 5.893 79.42 14.6 1.4
462 7.25 9.64 92.2 97.91 3.681 125.5 14.55 1.45
454 7.75 10.3 100.3 97.73 2.377 191.0 14.55 1.5
457 7.85 10.44 100.9 97.72 1.768 258.5 14.55 1.5
459 7.95 10.57 101.7 97.70 1.017 451.3 14.5 1.55
457 7.8 10.37 100.2 97.73 0.959 476.5 14.5 1.55
455 7.55 10.4 97.46 97.79 0.893 509.5 14.5 1.55
454 7.4 9.84 95.73 97.83 0.840 540.5 14.45 1.5
452 7.25 9.64 94.21 97.87 0.895 505.0 14.45 1.5
447 6.7 8.91 88.03 98.01 0.699 639.5 14.5 1.5
Helium in 12mm Tube, 1/29/1983
Pt lEl" l2 pb 0.1 (Eff)2 Vogum (p) 3 -Ls Lp
Watts (Watts) Watts - 8 cm W/cm cm cm
315 1.525 2.023 28.4 99.4 14.11 22.32 14.75 1.45
350 2.18 2.90 36.6 99.2 10.00 35.00 14.6 1.4
405 3.9 5.19 56.6 98.7 7.49 54.07 14.65 1.45
400 3.45 4.59 50.66 98.9 6.04 66.23 14.6 1.4
390 3.15 4.19 47.4 98.9 4.77 81.76 14.55 1.45
400 4 8 6.4 70.48 98.4 4.208 95.06 14.55 1.5
415 4.1 5.45 58.03 98.7 3.736 111.1 14.5 1.55
445 5.45 7.245 71.93 98.4 2.954 150.6 14.45 1.5
464 6 9 9.17 87.34 98.0 2.572 180.4 14.45 1.5
485 9 3 12.36 112.6 97.45 1.869 259.5 14.45 1.55
Oxygen in 12mm Tube, 2/3/1983
I"t (q, l2 Pb Qu (Eii)2 Vagume <p> 3 L5 L9
Watts (Watts) Watts - 8 cm W/cm cm cm
462. 10.25 13.63 130.3 97.05 4.77 96.86 14.7
444 7.75 10.3 102.5 97.68 3.66 121.4 14.6
402 9.5‘ 12.63 120.8 97.27 3.15 146.9 14.5
461 13 17.28 165.6 96.25 - - 14.45
457 17.75 23.6 228.1 94.84 3.14 145.4 14.45
470 20.75 27.59 259.3 94.13 2.016 233.1 14.45

40
72
108
148
196
257

474

“0

72

108
148
196
257
340
474

760
1100

30

72

108
148
196
257
340

89

Table 5.2 Experimental Results in 25mm Discharge Chambers

444
434
432
428
420
413
412
413
416
425

310
360

375
390
410
435
450
470
485
497

432
465
468
478
478.5
468.5
448

16.12
(mm)

5.4
5.05
4.9
4.65
4.3
4.05
3.9
3.8
4.05
4.3

15.12
(mm)
1.45
2.25
2.35
2.65
3.1
3.75
4.65
5.7

8.8
11.25

(Watts)
5.5

8.1
9.85
12.25
15.25
18.75
24.0

Nitrogen in 25mm Tube, 1/9/1983

’5

Watt!

7.18
6.71
6.51
6.18
5.72
5.38
5.18
5.05
5.38
5.7

00

71.44
68.4
66.6
63.8
60.1
57.6
55.6
54.0
57.2
59.4

(HT)
* 2

98.4
98.5
98.5
98.6
98.64
98.7
98.74
98.78
98.7
98.65

volume

12
7.8
3.7
2.5
2.0
1.96
1.96
2.34
2.4
1.96

<1»
li/o-3
35.5
55.5
115.5
111.4
255.1
215.95
259.5
115.4
112.4
215

Helium in 25mm tube, 5/1/1983

Pb cu
Wax: -
1.92 27.41
299 361
3J2 378
352 4L5
4J2 4669
439 537
6J8 628
168 144
93 875
1L1 1065
1496 1329
Oxygen in
01
1mm: -
LIN 14J8
1071 1023
1309 1226
1629 iﬂLs
20.21 187.2
24.93 235.1
3L91 M46

(Eff)
* 2

99.4
99.17
99.14
99.06
98.94
98.78
98.6
98.3
98.0
97.6
97.0

(Eff)
‘ 2

98.31
97.68
97.2

96.59
95.76
94.68
92.88

Vb use

23.3
23.4
19.0
15.1
10.9
7
4.8
3.815
3.26
2.58
2.1

Wimm

20.27
9.92
8.35
6.92
6.12
6.31
5.9

q»
lilo-3
13.3
15.4
19.16
24.8
35.8
58.8
90.0
118.0
144.0
187.6

25mm tube, 2/5/1983

q»
W/cmf
21.3
46.88
56.06
69.06
18.2
14.26
75.83

14.55
14.5
14.45
14.4
14.45
14.4
14.4
14.4
14.45
14.4

14.7
14.55
14.5
14.5
14.4
14.35
14.35
14.3
14.3
14.3
14.3

14.55
14.45
14.4
14.4
14.15
14.15
13.9

..—
‘D

e.e . e. e .a .
oommmqommm

...
13

e e e. a g
ﬁmmmohien

(l8

d-‘dd‘d‘jd-‘d
O C O O
U!
an

e e
Odo

40

72

108
148
196
257
340
474
584
160

Pressure

40

72

108
148
196
257
340
474
584
160

25
40
72
108
148
196
257
340

9(1

Table 5.3 Experimental Results in 37mm Discharge Chambers

Watts
431
433
427
414
406
400
400
410
420
450

8
Watts
310
325
350
370
385
403
424
447
462
477

Watts
412
430
448
410
481.8
490
481
418.5

lE.l2

(Watts)
1.18
1.53
2.0
2.525
2.915
3.55
4.35
5.55
6.45
8.05

IE,lz

(Watts)
3.9

5.4

6.6
8.65
10.5
13.25
16.25
19.0

Nitrogen in 37mm Tube, 1/9/1983

0u
Watts -
6.65 68.2
6.65 67.8
6.05 62.6
5.12 61.0
5.05 54.91
4.12 52.1
4.45 49.2
4.92 53.0
5.32 55.94
8.641 84.84
iiehiilJﬂl iii
Pb Qu
Watts -
1.51 22.36
2.03 21.65
2.66 33.6
3.351 40.0
3.955 45.38
4.12 51.1
5.18 60.26
1.38 12.92
8.515 82.0
10.1 99.12
Oxygen in
Watts -
5.185 55.6
1.18 13.16
8.17 86.5
11.50 108.1
13.96 128.0
11.61 158.8
21.6 196.0
25.26 233.2

(Eff)
* 2
98.74
98.33
98.04
97.55
97.1
96.41
95.56
94.72

Volume
cm
30.0
29.15
11.25
11.31
11.69
18.15
19.93
11.82

(Eff)2 volume (p)
8 cm W/cm3
98.46 24.32 11.12
98.46 11.06 39.15
98.58 9.259 46.12
98.62 6.416 64.52
98.76 5.887 68.97
98.82 4.913 80.43
98.89 4.18 95.16
98.8 3.318 121.4
98.13 4.562 92.01
98.08 4.549 98.93
37mm Tube, 1/9/1983
(Eff)2 Volume <p>
8 cm W/cm3
99.5 42.03 1.315
99.4 29.15 10.92
99.24 23.31 15.01
99.1 11.54 21.09
98.97 13.04 29.52
98.8 10.24 39.36
98.6 7.831 54.14
98.35 5.638 19.28
98.14 4.715 97.99
91.16 3.106 128.1

37mm Tube, 2/3/1983

(p)
W/cm3
13.12
14.15
25.91
27.06
21.23
21.0
24.43
26.85

on
14.55
14.5
14.45
14.45
14.45
14.4
14.4
14.4
14.4
14.4

cm
14.75
14.55
14.45
14.45
14.4
14.45
14.45
14.4
14.35
14.35

14.65
14.55
14.45
14.45
14.4

14.35
14.35
14.4

EU

_a__‘—.e_.—A._.—e_a—A_a06—
010101010101

cm
1.4
1.4
1.55

_a_.-—-I_.—D_a—‘
010101050101!-

9'1

Microwave Coupling Efficiency
Versus Pressure
for Nitrogen Discharges

 

 

 

 

 

 

 

 

100
N 99
' B ﬁe
a “ a % 9 g B B D
.33 98‘ O 00 8
“£3: __ O O O O O O
6
ca 97-
C Quartz Discharge Chamber
: -' Oriented Vertically
% 96 g 12mm i.d.
O "' 25mm i.d.
0 ._1 D 37mm i.d.

95 111111118 1 1111111

20 10
Pressure — Torr

Figure 5.2 Microwave Coupling Efficiency for Nitrogen

 

103

92

efficiency of over 97% was maintained for all three tubes and for a
pressure range of 40 to 800 Torr. Figure 5.3 shows the results for
helium. A coupling efficiency of over 97.58 was maintained for all
three tubes and in a pressure range of 40 to 800 Torr.

Using the equation for cavity Q described in chapter 4, the cavity
Q was calculated for nitrogen and helium in each of the three different
tubes. Figure 5.4 shows the Q for nitrogen discharges. The Q ranged
from 50 to 105 over a pressure range of 40 to 800 Torr in the three
tubes. Figure 5.5 shows the results of Q calculations for helium
discharges. The Q varied from 20 to 115 over a pressure range of 40 to
800 Torr for the three different tubes. The low values of Q compared to
the empty applicator case (about 5000) show that the majority of the
power is being absorbed in the discharge resulting in lower fields in
the cavity and thus lower wall loss.

Using the coupling efficiency, power absorbed, and discharge
volume the absorbed power density was calculated for nitrogen and
helium. Figure 5.6 shows these calculation for nitrogen. Values of 20
to 700 Watts/cm3 were calculated for a pressure range of 40 to 800 Torr.
Figure 5.7 shows these calculations for helium. Values of 7 to
300 Watts/cm3 were calculated for a pressure range of 40 to 800 Torr.
Figure 5.8 shows a comparison of these results of power density in the
discharge to earlier work by Rogers.13 This comparison is for 12 mm
tubes using the nitrogen and helium results described earlier in this
section compared to Rogers argon results. Rogers argon data was
produced using equipment identical to that used to produce the results
presented in this chapter. As can be seen in the figure, power density

went from 11 Watts/cm3 at 40 Torr to 120 N/cm3 at one atmosphere.

93

Microwave Coupling Efficiency
Versus Pressure
for Helium Discharges

 

 

 

 

 

 

 

 

 

100
_. g Q
N 5'
1%” Ogﬁgg
>. " O S
° 5
.5 98 - E
o
«“5 " Wat. °‘::.f'°12:.°".1'"b“ 0
en ca y
m 97 - 0 12mm i.d.
g ‘— D 25mm i.d.
:3- U 37mm i.d.
8 96-
o —l
95 111rr111 ‘1 1111111
20 108

Pressure — Torr

Figure 5.3 Microwave Coupling Efﬁciency for Helium

94

Cavity Q Versus Pressure
for Nitrogen Discharges

 

 

 

 

 

 

100- o o o O o o
._ O O
3 O
0 90— o
8 -‘ Quartz Discharge Chamber
*5 Oriented Vertically
D -— 0 12mm i.d.
“- 80 [> 25mm 1.4.
421‘ _ [:7 37mm i.d.
'6 70_ D
5 # D [E p
s ' a B
> 60- D p
8 D a
- [:1 D E
El
'50- [:1
11111111 1 1111111

 

 

 

102
Pressure — Torr

m
0

Figure 5.4 Cavity Q Versus Pressure for Nitrogen Discharges

Cavity Quality Factor, Qu

95

Cavity Q Versus Pressure
for Helium Discharges

 

 

 

 

 

 

 

 

 

120
.. Quartz Discharge Chamber 0
Oriented Vertically D
_ 0 12mm i.d.
100 [> 25mm i.d. D
"‘ 0 37mm i.d.
80 - B
60 -
O
_ o E B
40 - B
9 B
20 11111111 11111111
20 10a 103
Pressure - Torr

Figure 5.5 Cavity Q Versus Pressure for Helium Discharges

96

Power Density Versus Pressure
for Nitrogen Discharges

 

 

 

 

 

 

3
1o _4
—i Ouatz Discharge Chamber
_ Orlmtad Vertically
.... 0 12mm i.d. /0
_ D 25mm 1.41. ’00
d [:7 37mm i.d. f0
O

01
1

lllllj
O
A
‘3
El

Power Density — W/cm3
8 w
N

 

 

 

CID

D /

_ Cl
1) E]
-—+
" 1:1
10 11111111 1 1111111
20 1o2 103
Pressure — Torr

Figure 5.6 Power Density Versus Pressure for Nitrogen Discharges

97

Power Density Versus Pressure

500

0

1111111”

Power Density — W/cm3

O

lllll

for Helium Discharges

 

—

l

l

l

 

Quatz Discharge Chamber

Oriented Vertically

 

 

0 12mm i.d.
D 25mm i.d.

C] 37mm i.d.

 

 

 

 

20

llllllll l lllllll

102
Pressure — Torr

 

103

Figure 5.7 Power Density Versus Pressure for Helium Discharges

98

Power Density Versus Pressure for Nitrogen,

Helium and Argon in 12mm i.d.
Quartz Discharge Chambers

 

 

 

 

 

103g

: Cl — Nitrogen Cl
: o — Helium £1436
— [> — Argon“ E]

I")

E — (t—from Rogers“)

0

\

3 ..1

l

>.

+1 2

'(7, 10 :

s 1

D _.

E _—1

3 _.

o

D- “'1

lo 11111111 1 1111111
20 1a2 103
Pressure — Torr

Figure 5.8 Power Density Versus Pressure for Different Gasses

eriments for

of these exp
Figur

5.10 shows the cav'
difficult since the power

density vs.
arge at press

disch
limited to onl
However at all pressures

'scharge. It was possible

a evidence was found

oxygen but n
redone with a

y of more power.
ischarges in each of the different gases behaved the
tricted in the center

All of the d
5 increased,

pressure we

becoming cons
This occurred

mber walls.

same as the
Even thoug

from the cha
h the

chamber away

he discharge
ge chamber.

of t

regardless of the size of the dischar
such as

the tube size was

arc properties discussed in

power dens
changed. This conf1rms the descr1ption of
chapter 3. As the pressure 15 1ncreased, recombination increases,
Power coupling into the discharge is
Any movement

of the discharge chamber by design.
reduces energy coupling into the discharge and mismatches
Both of the diatomic gases, 02 and N2, showed consistent behavior

100

Microwave Coupling Efficiency Versus
Pressure for Oxygen Discharges

 

 

 

 

 

 

 

 

 

lOO
Qua'tz Discharge Chamber
Oﬁmtod Vertically
‘ 0 12.111.11.11.
D 25mm 1.8.
N 9 8 _. E] 37mm i.d.
I _l
5‘
.5 96 — DD 1:1 5
0
1‘3] - o B E
g. D
8 D D
o ‘ O
92 - D
90 11111111 1 1111111
20 1o2 103
Pressure — Torr

Figure 5.9 Microwave Coupling Efﬁciency for Oxygen

107

Cavity Q Versus Pressure
for Oxygen Discharges

 

 

 

 

 

 

 

 

 

350
Quartz Discharge Chamber
- 011.8188 Vertically
0 12mm i.d. D
300 - D 25mm m
C] 37mm1.d.
3 _
O: 250— O
jg _ o D 1:]
8
L1. 200'— 1:]
g _ D
:2 a
g 150 — D D
.2: ‘ O b D
'g 100— D D
o [:1
1 D [:1
50 — [3
O llllllll l lllllll
20 102
Pressure — Torr

Figure 5.10 Cavity Q Versus Pressure for Oxygen Discharges

1o3

Power Density -- W/cm3

102

Power Density Versus Pressure
for Oxygen Discharges

 

lO

 

Quartz Discharge Chamber
Oriented Vertically

0 12mm i.d.
D 25mm 1.8.

[:1 37mm i.d.

 

111111”

 

 

 

\

aix)
llllll
\.
A
‘71
1%

J

D—D—D—El-[j-D

l

V
Cl 1:]
10 11111111 1 1111111

20 102 103
Pressure — Torr

 

 

 

Figure 5.11 Power Density Versus Pressure for Oxygen Discharges

103
when compared to the monatomic gases Ar and He. Power densities were
higher and cavity Q was higher for the diatomic gases indicating

different loss processes than for monatomic gases.

5.4 SUMMARY

These experiments demonstrated the ability to generate non flowing
helium, nitrogen and oxygen at pressures of 40 to 1000 Torr. Over 95%
of the available microwave power was coupled into microwave discharge
over the entire pressure range. This figure could easily be improved by
the use of less lossy materials in the cavity walls and bettered
designed microwave coupling circuits. As the pressure was increased the
discharges contracted away from the discharge chamber wall forming small
volume discharges. All of these results are desirable properties for an

electrothermal thruster and serve to show that the concept is feasible.

CHAPTER VI

DEMONSTRATION OF A MICROWAVE ELECTROTHERMAL THRUSTER

6.1 INTRODUCTION

The design and testing of a microwave electrothermal thruster is
described in this chapter. The results of these experiments were the
first to demonstrate the feasibility of the microwave electrothermal
thruster concept.1 An experimental description of the thruster design
used for these experiments is presented first, along with a description
of how the experiment was conducted. Experimental measurements of
thrust and specific impulse were taken using the direct thrust
measurement system mentioned section 4.4.1. Efficiency, specific
impulse and power to thrust ratio calculations were computed from these

measurements and are presented in this chapter.

6. 2 M14511”; svsrgg
A description of the experimental equipment and technique used to
obtain the experimental data presented in this chapter is described in
this section. The microwave electrothermal thruster is described first.
Then a description of the experimental system used in measurements of
the properties of this thruster is presented. The design and
development iterations that produced the coaxial thruster are presented

last.

104

105

6.2.1 PROTOTYPE ELECTROTHERMAL MICROWAVE THRUSTER

The prototype electrothermal microwave thruster, used to generate
the results presented in this chapter, is shown in Figure 6.1. The
design of this thruster was complicated. Because of this, the
development of the thruster is detailed in section 6.2.3 showing the
design iterations that led to the successful testing of this thruster.
Figure 6.1 shows the entire thruster including the microwave coupling
structure. Figure 6.2 shows an enlargement of the discharge chamber
area of the thruster so details to fine to be shown in Figure 6.1 can be
seen. Figure 6.2 continues the numbering system used in Figure 6.1
where applicable.

As shown in the figures the thruster consists of a coaxially fed
microwave discharge located at the end of a 4.8 cm i.d. cylindrical,
brass microwave coupling system. These figures show how the thrust
discharge structure was incorporated onto the coaxial applicator. A
teflon plug (6) with O-rings, seals the discharge zone from the external
atmosphere air of the coaxial coupler (2), and provides a seal between
the vacuum and the plasma discharge chamber (8). The plasma discharge
chamber (8) is formed by a 23 mm i.d. quartz tube surrounding the
discharge zone and narrowing to a 1 mm diameter nozzle downstream from
the plasma and outside the electromagnetic excitation region. The
cylindrical screen (9), which forms the outer waveguide conductor for
the discharge chamber, is 6.5 cm in diameter and 9.5 cm long allowing
only evanescent electromagnetic wave and slow wave coupling to the
plasma discharges (3).

The input gas flow (7) is distributed around the teflon plug (6)

by a distribution ring and enters the plasma discharge chamber (8)

106

1 1
1-—6.5cm—-|

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Experimental
C o 0 x1 al
M 1 c ro w 0 ve
Thruster
L1
*—4
24.4cm
2—’ 41.7cm
1_. ___ M1- _____ _ _
L2
1 5/8" ElA ______ __
Legend Coax Connector -—5 61 lcm
(1)—Coaxial Microwave input 17 Sam
(2)-Coaxial Coupling Structure .
(3)-Microwave Discharge
(4)—Center Conductor
(5)—Sliding Short 1 l _______ 'l'_'__
(6)—Teﬂon Plug I El
(7)-Gas Input } :
(8)—Discharge Chamber 1 __|___ ________
(9)—Conducting Screen _L.: l|<-L-2.05cm o.d.
(lO)—Vacuum Flange —-£ 1 1 :-—4.8cm i.d.

Figure 6.1 Experimental Coaxial Microwave Thruster

107

Thruster Discharge Chamber

 

 

 

 

 

 

 

 

 

\\\\\\
7 —" *1.) U

.. 4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E‘s—- O—Rings W“ Gas Flow

(For Legend see Figure 6.1)

Figure 6.2 Enlargement of Thruster Discharge Chamber

108

through holes drilled at angles across the center axis of the plug.

This produces a vortex flow in the chamber helping to stabilizing the
discharge and causing a vortex flow through the discharge. Figure 6.2
shows the approximate gas flow pattern around the teflon plug and
through the discharge chamber. At high pressures (>100 Torr) the
flowing gas produces a stable discharge constricted away from the quartz
walls of the discharge chamber, but attached to the end of the center
conductor (4). The center conductor is water cooled to prevent melting
of the inconel tip. A vacuum flange (10) is use to attach the thruster
to a vacuum chamber. Figure 6.2 shows the O-rings used to seal the

chamber in more detail.

6.2.2 MEASUREMENT SYSTEM AND EXPERIMENTAL PROCEDURE

The input microwave source was a continuously variable,
200-600 Watt, cw, 2.45 0H2 magnetron oscillator. The input microwave
system consisted of the same measurement system described earlier in
section 4.2.2 and provided direct measurement of incident and reflected
power. Microwave power is matched to the discharge by adjusting the
sliding short, (5) in Figure 6.1, and by adjusting the center conductor,
(4), for minimum reflected power. The important electrical lengths are
shown in Figure 6.1 for this coaxial applicator. Lengths L1 and L2
indicate the center conductor and sliding short positions for a given
experiment.

The experimental thrust test were performed at NASA Lewis Research
Center's vacuum tank 88 using the experimental setup in Figure 6.3. As
shown in this Figure the thruster was attached to the vacuum tank by 4"

vacuum flange, (10) in Figure 6.1. This allowed the thruster to be

109

EXperimental Thruster Setup

 

\?
/

r—Torsian Head

-— Music Wire
— Thrust Target --—Guide Tube

—— Vacuum Tank

 

 

 

 

 

Gear Mater——

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E — .—
ll
4" Vacuum
Flange I
Center Line
Experimental Coaxial faxfszaanv:

Microwave Thruster Thrust Target

k L— Pre—Load TorqueX Head

,3
I

 

 

 

 

 

Figure 6.3 Experimental Setup of Microwave Thruster

110

exhausted into a vacuum environment held to below 10" Torr, assuring
that a choked flow condition exists. Exhaust gas from the thruster
impinges against a flat target of the direct thrust stand described in
section 4.4.1. As shown in Figure 6.3, this thrust stand is mounted in
the tank to allow the direct reading.

Experimental measurements were carried out using the experimental.
flow and pressure measurement system described is section 4.2.1.
Nitrogen gas was tested at discharge pressures of 100 Torr to over one
atmosphere and with flow rates in the range of 6.4x10‘5 kg/s to

11.'7x10'5 kg/s.

6.2.3 COAXIAL THRUSTER DEVELOPMENT

Figure 6.4a shows the initial applicator thruster design that was
tested as a thruster concept. The applicator continues the numbering
system used in Figure 6.1. The tests with this thruster in nitrogen gas
were unsuccessful due an inability to keep the discharge on the tip of
the center conductor (4). As the pressure was raised the discharge
would form against the teflon plug (6) and melt it.

The thruster was redesigned and is shown in Figure 6.4b. The
discharge chamber (8) was shortened and the teflon plug (6) was moved
closer to the end of the torch. The thruster was set up and tested in a
vacuum tank with a thrust stand to measure thrust. The thruster worked
so well that even moderate levels of input power exceeded the thrust the
thrust stand could measure. Some difficulty in tuning the thruster was
encountered, so while the thrust stand was modified to measure greater
thrust, the thruster was slightly changed to solve this.

The thruster was changed to the design shown in Figure 6.1. The

Experimental

Coaxial

Microwave

Thruster

Designs 9—-
110

7—.
T
1__.
Legend

 

 

(1)-Coax” Microwave Input
(2)-Coax” Coupling Structure
(3)-Microwave Discharge
(4)—Center Conductor
(5)-51181ng 910d

(6)-Teﬂon Plug

(7)-Gee input

(8)-Discharge Chamber
(9)-Conducting Screen
(10)-Vacuum Flange

><

/
(A

111

 

 

 

 

 

 

 

 

 

llll

 

 

(0)

1.__.

 

 

E3224

 

 

 

 

ill

(b)

Figure 6.4 Experimental Microwave Thruster Design Variations

112
major change was to the shape of the screen outer conductor (9).
Experiments with this thruster using nitrogen gas were very successful
and are presented in this chapter. Experiments were also conducted with
helium, but low efficiencies were observed (under 10%). It was felt
that a discharge chamber with a smaller nozzle would solve this problem.
A center conductor that could feed gas out the tip was also tested. It
was felt that the efficiency could be improved if the discharge was
moved off the tip by flowing gas through this tip. In experiments even
the slightest gas flows would blow the discharge out and were unable to
move it off the tip and still maintain it.

The tuning and matching of this thruster continued to be a
problem. The best match that could be obtained was 85% but earlier
experiments with the applicator shown in Figure 3.4 were able to match
95* or better into the applicator. The larger outer diameter of the
screen (9) was felt to be causing a electromagnetic mismatch at the
boundary with the outer conductor (2) of the coupling structure. The
screen was modified and is shown in Figure 6.5a. With no change in
diameter, the impedance to the discharge region would remain the same.
However the efficiency of the thruster dropped and the tuning became
more difficult and the modification was dropped.

Tests with hydrogen were desirable but the quartz presented
difficulties. Chemical reactions between the hydrogen and the quartz
would ruin the nozzle. A metal wall discharge chamber shown in Figure
6.5b was constructed. This thruster was tested in hydrogen and
efficiencies of ~10: was observed. However hydrogen reactions with the

teflon plug (6) severely damaged it and the tests were discontinued.

Experimental
Coaxial

Microwave 9__.

Thruster
Designs 10
7—.
1_.
Legend

 

 

(1)-Coaxial Microwave input
(2)-Could Coupling Structure
(3)-Microwave Discharge
(4)-Center Conductor
(Sl-Sldhg Short

(6)—Teﬂon Plug

(7)-Gee input

(3)-01mm Dunbar
(8)-Conducting Sore-1
(10)-Veeuum Flange

 

 

113

 
 
 
 
   

 

-—5

 

 

 

ill
(a)

 

 

 

 

 

 

 

ill
(b)

Figure 6.5 Microwave Thruster Design Variations

114
6.3 EXPERIMENTAL MEASUREMENTS

Experimental results presented in this chapter were produced with
the thruster shown in Figure 6.1. Figure 6.6 shows the thruster working
in vacuum tank #8. It is operating with a input power of 464 N and a
flow rate of 11.'7x10"5 kg/s. The discharge size increases with input
power and decreases with increases in pressure and flow rate. Typical
discharge lengths and diameters increase from 1 to 4 cm and from 1 to
1.8 cm respectively as power is increased from 200 to 600 Watts.

The measured experimental performance of the thruster for
different nitrogen flow rates is shown in Figure 6.1. All of these
results are for the thruster operating in a steady state mode. The
thruster was operated at times for periods of over 3 hours. No erosion
or melting was observed during this time while using nitrogen gas as a
propellant.

The specific impulse, defined in section 2.3.1 and equation 2.1 as
the ratio of thrust to the input propellant flow rate, is plotted versus
measured thrust in Figure 6.7a for the fixed nozzle size of 1 mm. Each
curve represents thrust measurements made with constant gas flow but
increasing absorbed power from 200 to 600 Watts. Since the nozzle size
was fixed, each level of propellant flow established a corresponding
range of discharge pressures. This pressure variation along with the
corresponding variation of input power is indicated in the figures
legend. Increases in absorbed power resulted in increases in measured
discharge pressure and thrust.

The cold specific impulse (specific impulse without a discharge)
is shown in the lower left hand corner of each curve in Figure 6.1a, and

is nearly independent of gas flow rate and agrees well with theoretical

115

Picture of Operating Microwave Thruster
at Lewis Research Centers Tank #8

Mass Flow=10.6x10—5kg/s input Power=474 Watts
Discharge Pressure=800 Torr

 

Figure 6.6 Picture of 0 Operating Microwave Electrothermal Thruster

Specific impulse — Seconds

Energy Efﬁciency - X

116

Thruster Perform an ce

 

 

 

 

 

 

 

 

300
‘ O/ / /
D /A
200 — /°O OFF/@330“
.. /’ /D , //O
100 ”/” ”//’
O/D A’O’ Mass input Discharge
.1 Point Flow Power Pressure
( O ) x1 O-aig/s Watts Torr
O — o 6.4 344-490 438—493
D 8.5 195-480 460-666
A 10.6 240-474 BSD—800
100 _ O\ D\ Ai O 11.7 200-464 744-923
80 — x x \
.. \\ \ \\\
60 — ‘ \ 9\ E
_ \ \ <9 (953‘
\\ UB4] 1:]
40 -l \ D
O
- OO\
20 -‘
d (b)
0 l r 1 I T j l I i r 1 I
0 4 8 12 16 20 84

Thrust - Newtons x 10"2

Figure 6.7 Coaxial Microwave Electrothermal Thruster Performance

117

calculations of room temperature gas. From its definition, equations
4.6 and 4.7, the specific impulse should have a straight line
relationship versus thrust and a slope that varies as the reciprocal of
the propellant flow rate. The experimental points in Figure 6.7a
demonstrate this relationship. A straight line drawn through the "hot"
data points is extrapolated to each corresponding cold flow point.
Extrapolated extensions of these lines pass through the origin as
expected.

The energy overall efficiencies are shown in Figure 6.7b. The
efficiency is defined using the equation 4.9 in chapter 4 and the data
from Figure 6.1a. As shown in Figure 6.7b, efficiency improved with
propellant flow rate, reaching a maximum of about 60%. There is a large
gap in points between cold and hot efficiencies since the microwave
generator did not produce regulated power below about 200 Watts.
Furthermore it is doubtful that the plasma discharge is stable at lower
power levels. Over the absorbed power range of 200 to 600 Watts and at
a constant flow rate, the efficiency decreased 10% or less with
increasing power.

The power to thrust ratio plotted against specific impulse is
shown in Figure 6.8. Lines of constant efficiency are shown, but they
are terminated at the assumed cold specific impulse value of 80 seconds,
because at that point all values of power to thrust ratio go to zero.
It is interesting to note that although the power to force ratio is
identically zero at unity efficiency when hot specific impulse is equal
to the cold specific impulse, other values of hot specific impulse at

near 1003 efficiency result in a finite power to thrust ratio

Watts

 

~

 

Power

118

Power to Thrust Ratio for the

 

 

 

 

 

 

 

Microwave Electrothermal Thruster
1111:: 1.2:": 1:382: L1 L2
x10-5Ttg/s Watts Torr mm mm
o 6.4 344—490 438-493 275 :57
1:1 8.5 195—480 480-666 269 38
A 10.5 240—474 630-800 275 37
o 11.7 200—464 744—923 - —
4000-— / /
_. O //
/ / /
(D O / /
g 3000— / / // /
g a O / ,/ / /’
Z _ / /D / // //
1:1 /
44 2000.. 1 [,13/ A ////
U) S C / / ////
E -1 F // I] /// /’//
1"" i / //’
1 ’////
1000— 1/ ,/0/ , x’g-to
/ // / //// \K\(\q
1 ”I
O 1 1 1
D lOO ZOO 3100 400
Specific Impulse —- Seconds

Figure 6.8 Power to Thrust Ratio for the Microwave Thruster

119

2
proportional to the quantity IspH - Ispc . When IspH>>Ispc' this
IspH
quantity approached IspH' such that the power to force ratio approaches
gIspH . Using the relationship IspH'xﬂ , the power to thrust ratio
2 9

then approaches 25 , which is also the time rate with which kinetic
energy is impartgd to the propellant exhaust divided by the thrust
produced. Under the best of conditions the power to thrust ratio cannot
be less than one half the exhaust velocity (in consistent units). At

lower values of IspH' the power to thrust ratios are calculated for

efficiency equal to 1003 and are drawn in Figure 6.8 as a broken line.

6.4 COMPARISON OEﬁEXPERIMENTAL THRusrggs

A performance comparison of the microwave electrothermal thruster
against the data from two resistojetszor21 and an arcjet25 from chapter
2 is shown in Figure 6.9. The resistojets, designated A20 and 821,
differ in heating approach and consequently in the dimensions of the
propellant passages. The arcjet used in the comparison , C25, was a 1
kW system.

The experimental results prove that a working microwave
electrothermal thruster can be built. As shown in Figure 6.9 the
performance of the microwave electrothermal thruster compares favorably
with other electrothermal concepts Operating in nitrogen gas. They also
show that this type of a thruster can produce a significant increase in
thrust over cold flow and that the efficiency of the input energy to
. output thrust can be as much as 608. It should be pointed out that

these results are only the most preliminary since this thruster was

120

Comparison of Several
Electrothermal Thrusters with the
Microwave Electrothermal

Thruster
A.B - Resistalet C - Arcjet D - Microwave
Electrothermal Electrothermal Electrothermal
Thruster Thruster Thruster

100

 

Propellant

"" LL91} -N2
62 80 a O-NHa
ml- @1151!) 0 -He

FIE]
8.1216:

\o\o 113-H2
911..o__ .4

it}; [9%

L

a
o
I

 

l
/

Energy Efﬁciency
6
J

 

 

 

 

 

20 — , D am
4 :m «HE “5'
o l I 1 T r I 1
o 200 400 600 800

Speciﬁc impulse — Seconds

Thrust in Newtons

Electrothermal E
Thruster —_l

Propellant Flow Rate in 10" Kg/s

 

 

Figure 6.9 Electrotherrnai Thruster Comparison

121
built as a prototype and was not yet optimized to increase performance.
Hopefully experiments with this thruster will continue and show greater
improvements. It would be desirable to test H2 and NH3, and run tests

at higher power and chamber pressure.

CHAPTER VII

CAVITY APPLICATOR THRUSTER RESULTS

USING QUARTZ NOZZLES

7.1 INTRODUCTION

Experiments in Chapter 5 demonstrated the ability to generate and
maintain a high pressure microwave discharge with a cavity applicator
and with no gas flow. These experiments showed that microwave energy
could be coupled into the discharge with a efficiency of greater then
983 and that power densities of greater then 100 W/cm3 could be
achieved. This chapter continues these experiments by applying the
knowledge gained from chapter 5 to the development of a cavity
applicator based microwave electrothermal thruster. While the prime
emphasis of this chapter is on development of a thruster, the results
presented and technique described can be extended to any microwave
cavity applicator based discharge using a flowing gas. Indeed, while
demonstrating a working thruster, this chapter also demonstrates the
ability to generate and maintain high pressure microwave discharges in a
flowing gas environment.

This chapter presents results of experiments using a thruster,
incorporating a cavity type applicator for energy coupling. An
experimental description of the thruster and the setup used to test it
is presented first. Experimental measurements of thrust and specific

impulse are presented next. Efficiency and power to thrust ratio are

122

123

derived from the thrust measurements.

7 .2 QLEERIMENTAL svsg

An experimental thruster was built based on the design of the
applicator shown in Figure 3.5. The thruster based on this applicator
is shown in Figure 7.1. This figure continues the numbering system used
in Figure 3.5 with the microcoax system omitted from the drawing to
simplify it. The cavity applicator from Figure 3.5 has been modified to
include a discharge chamber (6) with a nozzle (13) to convert the
thermal energy into thrust and is shown as (D) in Figure 3.7. For all
experiments presented in this chapter a cylindrical quartz discharge
chamber was used with inside diameter of 23 mm or 28 mm and forced air
cooling was added to prevent the chamber from melting. The nozzle was
located at the exit of the cavity, as shown in Figure 7.1, for all the
measurements of this thruster. Also a adjustable center body (14) has
been added to help stabilize the discharge (4). The variable length
input probe (8) and the sliding short (2) used the actuators described
in section 3.4.1 (and shown as (B) and (C) in Figures 3.6 and 3.7) for
more accuracy in the measurements.

The flow and pressure measurement system described in section
4.2.1 and shown in Figure 4.1 was used for the measurements in this
chapter although only one gas was used at a time. The applicator was
oriented in a vertical manner with the gas flow downward and out the
nozzle at the bottom of the applicator. The small size of the vacuum
system made it necessary to use a heat exchanger to cool the exhaust gas
and prevent damage to the pumping system. These experiments were set up

at Michigan State University and used a roughing pump to exhaust the

124

Cross Section of a Cavity Applicator
Based Microwave Electrothermal

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Thruster
Legend .1 l ...
(1)-Cavity Walls 1
(2)-Sliding Short ; :1 E :3
(3)-Base Plate
(4)-Plasma Discharge
(5)-Viewing Window . . . .
(6)-Discharge Chamber 1 :3 E In
(7)—Coaxial Microwave input
(8)-input Coupling Probe
(9)-Brass Collars 1
(13)-Nozzle
(133-Nozzle (see text) ‘—J
(14)-Center Body +__1 4
.8 T U 1 .
C. I L__ 1__:I n
2 6—.-
9 L8
(.39 4—I1l11
Tj’ D \ .14 5
_..‘ 1— ‘8 { —"
7 -1" 1 13* 7) \K ‘
T LA I— _7_] ll
L i
131—- 3

Figure 7.1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Cross Section of a Cavity Microwave Thruster

125
propellant. The microwave supply system of section 4.2.2 was used for
measurement and control of microwave power. For these experiments a
continuous wave, variable power, 0 to 2500 Watt microwave power source
was used.

Experimental runs were performed by first establishing a desired
propellant flow rate and holding this flow rate constant throughout the
entire experimental run. Once the desired flow rate was achieved, the
cold discharge pressure was measured. The discharge was then ignited
and the input microwave power was adjusted to the desired level. The
applicator input power, Ptv and the steady state discharge pressure pH
were measured for several operating points as the input power, Pt' was
varied. After each experimental run, the discharge chamber was allowed
to cool back to the initial conditions as a check against nozzle
degradation.

The thrust force, specific impulse, energy efficiency and the
power to force ratio were calculated from equations 2.1-2.5 and 4.8-4.9
with the thrust force being derived using the indirect thrust method
described in section 4.4.3. As discussed in chapter 5, microwave
coupling efficiency was very high, therefore Pt was assumed to be equal
to Pa for these calculations. The input power Pt was increased until
nozzle heating threatened to destroy the nozzle. Thus all experiments
were limited by nozzle heating.

Results of experiments performed by S. Nakanishi6 of NASA Lewis
Research Center are also presented in this chapter. These experiments
used an experimental setup and technique nearly identical to that
described above. For these experiments a 0 to 600 Watt source was used

with a cavity applicator identical to that described in Figure 7.1.

126

However in these experiments the position of the nozzle could be varied
with respect to the position of the discharge. For the results of
experiments presented, the nozzle was moved from (13) to inside the
cavity (13*). A large vacuum tank with a diffusion pumping system was
used to exhaust the propellant. The size of this system and its pumping
speed made a heat exchanger, shown in Figure 4.1, unnecessary. This
experiment was also oriented vertically, however in this experiment the
gas flowed up through the applicator and the through the nozzle at the
top. This orientation was felt to be beneficial since the gravitational
gradient and buoyancy would increase the flow of heated propellant to
the nozzle. Experiments were performed using this system without

external forced air cooling of the discharge chamber.

1.3 EXPERIMENTAL MEASUREMENTS

Initial experiments were performed with nitrogen in a single
applicator TM012 mode to develop a basic understanding and technique in
thruster operation. The first section presents the results from these
experiments. Further experiments were conducted with nitrogen in a
TM011 mode to study the affect the mode has on the thruster operation.
The next section presents the results of these experiments and the last

section presents a review of these results.

7.3.1 EXPERIMENTS WITH NITROGEN IN DIFFERENT MODES

Initial experiments were performed with nitrogen in a TM012 mode
using a 23 mm quartz discharge chamber with a nozzle diameter of 1.5 mm.
The position of the discharge and the pattern of the electromagnetic

fields are shown in Figure 3.8. The results of these experiments are

127

shown in Figure 7.2 as energy efficiency versus input power for flow
rates of 63x10"6 kg/s to 146x10"6 kg/s (3000 sccm to 7000 sccm) and the
experimental range of conditions are described in Table 7.1 as paint 1,
2 and 3. These data points proved difficult to maintain at higher input
powers and at higher flow rates. The discharge would become unstable in
position and sometimes be swept downstream from the intense microwave
region and extinguished. However the thruster was operated for periods
of greater than 1 hour with no nozzle melting or erosion, but had
problems with the quartz discharge chamber deforming next to the
discharge.

Further experiments were conducted using a discharge chamber with
the stabilizing center body (14) shown in Figure 7.1. The data
corresponding to these experiments is shown in Figure 7.2 and is
described as points 4, 5 and 6 in Table 7.1. These experiments used a
quartz discharge chamber with a inner diameter of 23 mm and nozzle
diameter of 1.17 mm. The stabilizing center body served a dual purpose
of forcing a boundary layer flow along the inner surface of the
discharge chamber and of providing a wake or recirculation zone similar
to a combustion stream flame holder. The center body worked as planed
producing stable discharges. However, as the input power was increased
the discharge extended to the nozzle and produced problems with nozzle
melting. Thus the power levels were limited to < 2000 Watts to prevent
the nozzle from overheating.

Additional experiments were conducted using nitrogen with the
BPPilCBtOP operating in a TM011 mode using a 28 mm diameter discharge
chamber with a nozzle diameter of 1.17 mm. The position of the

discharge and the pattern of the electromagnetic fields are shown in

128

Table 7.1 Operating Conditions for Experiments with Nitrogen Propellant

Operating Conditions for
Experiments with the
Microwave Thruster of Figure 7.1
in Nitrogen Propellant

 

F 101111 Discharge Cavity Nozzle Center L5

5 Point Rate Pressure Mode Size Body LP

(kg/sXIO‘é) (Torr) cn cm
1. 0 63.0 258-306 111013 1.5mm no 14.55 1.95
2. V 104 440-486 T114012 1.5mm no 14.5 1.95
3. 0 146 568-624 1111012 1.5mm no 14.5 1.95
4. 0 63.0 503-525 TMOIE 1.1711111 yes 14.5 1.95
5. I 104 710-819 TMOIZ 1.1711111 yes 14.5 1.95
6. A 146 953-1081 TM 013 1.171'17'1 yes 14.5 1.95
7. f 36.5 380-440 T111011 1.1711111 no 7.1 0.9
8. O 146 882-1000 TMOII 1.1711111 yes 7.1 0.8

W

Energy Efficiency 7.;

129

Energy Efficiency Versus
input Power
for Operating Conditions

 

 

 

 

of Table 7.1
50
40—
0' \
30— Q
\
. A
\A\‘
80_ ‘\ IM~’
K. eliL—J’V
10 — Fe—e—e
0 1 1 1 ' 1 1

0 400 800 1200 1600 2000

Power — Watts

Figure 7.2 Energy Efﬁciency Versus input Power for Nitrogen

130
Figure 3.8. The results of these experiments are shown in Figure 7.2 as
solid points and are described as points 7 and 8 in Table 7.1. The
discharge was next to but not quite touching the nozzle for both points
using the TM011 mode. For data point 7 no flow stabilization was used,
limiting the maximum flow that discharge could be sustained at. Even at
the flow rate shown for point 7 there was still problems with the
discharge similar to that in the TM012 mode with no stabilization, with
the discharge being unstable and easily extinguished. Data point 8 used
the same-flow stabilization system described earlier. Stability was.
excellent at high flow rate, however the same nozzle melting problem
mentioned earlier limited input power to less than 1 kW.

Figure 7.3 shows energy efficiency versus input power for
experiments conducted by S. Nakanishi at Lewis Research Center. For
these experiments a 29 mm i.d. quartz discharge chamber with a center
body stabilization system was used with a 1.27 mm nozzle. The nozzle
position in this system could be changed for different experiments, thus
changing the distance from the discharge to the nozzle. Figure 7.3
shows the results for two different positions under identical condition
of input power and propellant flow. It was observed that the best
performance, i.e. energy efficiency and specific impulse, was obtained
with the nozzle located inside the cavity from 3.6 to 3.0 cm from the
fixed end plate. This distance is approximately one quarter
electromagnetic wave length for the TM012 mode inside the cavity and is
located at an axial electric field minimum. Thus the nozzle is located
at the end of the plasma discharge core.

Specific impulse versus energy efficiency is presented in

Figure 7.4 for all the points in Table 7.1 and the nozzle position

131

Energy Efficien cy Versus
Input Power

for Experiments
by S. Nakanishi

(see reference #6)

 

 

50
40 -
N
5. %
.6 30— g
.2
B]
>. 80 -
9
a
c
“J 10 —
Mass Flow = 94x10”6 kg/s
0 1 1 1 1

 

 

0 200 400 600 800 1000

Power — Watts

E —Nozzie Position 3.74cm inside Cavity
® -—Nozzie Position 3.59cm inside Cavity

Figure 7.3 Results of EXperiments by S. Nakanishi

132

Energy Efficiency Versus
Specific impulse
for Operating Conditions
of Table 7.1

®~ See Figure 7.3

 

 

 

50
N 40“ ®
E O%®
.92 30- e
1.9.-
1:] . ‘A‘
a 20 "' .‘I I‘d
&
g ‘0. 0‘ 4‘
“J 10- In!
0 1 1 1 1 1 1
0 100 800 300

Specific impulse - Seconds

Figure 7.4 Energy Efficiency Versus Specific impulse far Nitrogen

 

133

experiments of Figure 7.3. Maximum specific impulse was approximately
280 sec corresponding to a cold to hot discharge chamber pressure ratio
increase of over three. The lower efficiencies for the points from
Table 7.1, were due in part to forced air cooling, smaller chamber

diameter and nonoptimum position of the nozzle.

7.3.2 EXPERIMENTS WITH HELIUM IN DIFFERENT MODES

The results in the previous section show that a cavity thruster
using nitrogen propellant is a valid concept for electrothermal
propulsion. However better performance was desirable along with
operations using lighter propellants. This section presents the results
of experiments using helium as propellant in the same cavity type
thruster of the last section. Helium was tested using two different
cavity modes, TM012 and TM011. Figures 7.5 shows efficiency versus
input power for the data point in Table 7.2. The legend in Table 7.2
details the different experimental configurations and operating
conditions under which the experimental points were taken.

Discharge behavior in the TM012 mode was similar to that observed
in the previous sections and is represented by data point 1 in Table
7.2. A 28 mm i.d. discharge chamber was used with a 1.17 mm nozzle and
stabilization apparatus. No problems with stability were encountered
but nozzle heating limited the maximum input power to less than 1.2 kW.
Lower flow rates were also tested but produced severe nozzle heating and
were discontinued.

Experiments With the TM011 mode with helium produced a discharge
that was next to but not quite touching the nozzle as in nitrogen. Data

points 2, 3, 4 and 5 from Table 7.2 used the same discharge chamber as

134

Table 7.2 Operating Conditions for EXperiments with Helium Propellant

Operating Conditions for
Experiments with the
Microwave Thruster of Figure 7.1
in Helium Propellant

 

F low Discharge Cavity Nozzle Center L L
‘7 Point Rate Pressure Mode Size Body 5 P

 

(kg/SXIDT6) (Torr) (:11 cm
1. 0 26.7 280-328 TMOIB 1.171111 no 14.85 1.75
2. U 8.9 170-196 TMou 1.171111 no 7.2 0.8
3. A 14.8 260-380 TMDII 1.1711111 no 7.2 0.8
4. O 20.8 330-430 TMou 1.171'17'1 no 7.2 0.8
5. <1 26.7 402-546 TMOII 1.1711111 no 7.2 0.8
6. 0 8.9 567 TMOII 0.511111 no 7.1 1.9
7. V , 10.4 685 TM 011 0.511111 no 7.1 1.9
8. 0 11.9 810 TM 011 0.511111 no 7.1 1.9

Energy Efﬁciency 7.

00
o
l

N
o
l

135

Energy Efficiency Versus
Input Power

for Operating Conditions
of Table 7.2

01
c3

 

.p.
o
J

21?;
Q9

1 l l l l
0 200 400 600 800 1000 1800

Power — Watts

 

 

 

Figure 7.5 Energy Efficiency Versus Input Power for Helium

136

was used for point 1. No stabilization problem were encountered and the

thruster was able to operate at several different flow rates. Nozzle

heating was still a problem and input power was limited to below 800 H.

As can be seen in Figure 7.5 efficiencies of up to 458 were achieved for

input powers of up to 750 Watts.

Addition experiments were conducted with helium in a TH011 mode

w‘i th a discharge chamber of 28 mm i.d. and a 0.51 mm nozzle. Data

points 6, 7 and 8 from table 7.2 represent these experiments. No

stabilization problem were encountered and the thruster was able to
Operate at several different low flow rates with good efficiencies.

Nozzle heating was a severe problem and input power was limited to below

500 N. This also limited the number of data point since the minimum

power required to sustain the discharge was near 500 H. As can be seen

in Figure 7.5 efficiencies of up to 408 were achieved for input powers

of up to 500 Watts. Figure 1.6 shows the energy efficiency versus

Specific impulse for all the points in Table 7.2. A maximum specific

impulse of 600 sec at a efficiency of 408 was achieved.

7 . 3.3 DISCUSSION

Discharge behavior was similar to that observed in the coaxial

thruster presented in chapter 6. For both modes the discharge lengths

and diameters increased as input power increased, and decreased or

Constricted slightly as pressure or flow rate increased. A comparison

be‘tmeen Figures 7.2 and 7.4, and the one for the coaxial thruster,
Figures 6.7 and 6.8, indicate that the experimental trends for the two
tht‘uster geometers are similar except for the larger thrust output in

the cavity thruster associated with its higher input powers.

137

Energy Efficiency Versus
Specific Impulse
for Operating Conditions

 

 

of Table 7.2
50
ﬂo\q
N 40— \Q 8
>
g A
,9; 30d \
0
SE
Lu 20 A
> " 1:1
0'0
as \U
C
L“ 10 a
0 l l ' i l 1 l

 

 

0 100 200 300 400 500 600 700

Specific Impulse — Seconds

Figure 7.6 Energy Efficiency Versus Speciﬁc impulse for Helium

138

In all the energy efficiency versus input power figures for a
fixed experimental geometry and for constant flow rate, the energy
efficiency decreased slightly as the absorbed power increased. Energy
efficiency increased with increased flow and increased in the presence
of a discharge stabilizing center body, while the thrust increased as
input power increased. As expected, specific impulse increased as
nitrogen gas mass flow rate decreased.

Figure 7.7 presents a comparison of specific impulse versus energy
efficiency for several data point from Tables 7.1 and 7.2. As can be
seen from the data, helium gas excited with the TM011 mode has higher
energy efficiencies than in the TM012 mode. In nitrogen gas this was
reversed and the TM012 mode had better efficiencies. Nitrogen gas had a
maximum specific impulse of 280 sec and helium had a maximum specific
impulse of up to 600 sec. Both of these result represent a increase of
about 3 times over the cold specific impulse and the higher helium
specific impulse can be attributed to its higher cold specific impulse.

Figure 7.8 presents a comparison of specific impulse versus thrust
to power ratio for several data point from Tables 7.1 and 7.2. As can
be seen from the data, helium gas excited with the TM011 mode has higher
specific impulse than nitrogen gas in the TM012 mode. However the
nitrogen has better power to thrust ratio. For all gases and modes the
specific impulse drops as the thrust to power ratio increases as
expected.

The results of these experiments show that a microwave
electrothermal thruster based on a cavity is a valid concept. Optimum
result were obtained in nitrogen in a TM012 mode while for helium the

best result were obtained for a TM011 mode. From the standpoint of the

En erg y Efﬁciency 7.

139

Energy Efficiency Versus
Specific Impulse

for Operating Conditions

of Tables 7.1 and 7.2

 

 

50
<i<>\<]
40 - _\ 8
Q,
Q A

30 - §

\

O
20 A

_ 1 1::
El' \ \
g‘d D

10 4
0 l 1 1 r I I

 

 

0 100 200 300 400 500 600 700
Speciﬁc impulse — Seconds

Figure 7.7 Energy Efficiency Versus Specific Impulse Comparison

140

Specific Impulse Versus
Thrust to Power Ratio

for Operating Conditions
of Tables 7.1 and 7.2

 

 

 

 

103

-3 - oo

8 v

o —

o

... “\:\.

I —

a

U)

s I— ‘\

Q.

g Me

0 N

.‘E ..

O

o km

a.

m

102 I I I 11 1 l l

0.05 0.1 0.5
Thrust to Power Ratio

Newtons/kWatts

Figure 7.8 Speciﬁc Impulse Versus Thrust to Power Ratio Comparison

141

electromagnetic field pattern, the TN011 is half a TM012 mode. Further,
from the discussion in section 3.5 on energy paths, the electromagnetic
coupling into the discharge should be independent of the electromagnetic
mode. The major difference between the two modes is the position of the
discharge relative to the nozzle. This implies that the distance from
the nozzle to the discharge is very important and the results of

S. Nakanishi presented in Figures 7.3 and 7.4 confirm this. Further
experiments need to be conducted that show what distance is optimal.

The experiments in nitrogen also showed the importance of gas flow
paths around the discharge. The addition of a center body improved the
operation and increased the performance of the thruster. Additional
studies need to be done to determine the optimal flow path for the
propellant. Nozzle melting limited all the experiments in the maximum
power they could handle. Further experiments in helium with very small
(0.51 mm) nozzles generated high specific impulse values but with very
severe input power limitations. Nozzle designs that allow for very
small nozzles for potential use with hydrogen are desirable. These
nozzles need to be designed to withstand greater temperatures and

chemical reaction problems associated with gases such as hydrogen.

CHAPTER VIII

EXPERIMENTS HITH METAL NOZZLE THRUSTERS

8.1 INTRODUCTION

The design and initial test of a revised version of the cavity
thruster is described in this chapter. Specific revisions are; (1) the
use of stainless steel nozzles, (2) quartz discharge chamber redesign,
and (3) operation over a wider range of discharge pressure. The
experimental measurements of thrust, specific impulse and energy
efficiency for different gas flow rates in nitrogen and helium gas are
presented. Measured performance is compared to the earlier cavity
applicator based microwave electrothermal thrusters presented in
Chapter 7. The experimental results in this chapter are described as
initial because at the time this thesis is being written, these are the
results that have been obtained. Further research with this thruster is
planed at Michigan State University, and will be the subject of future

papers.

8.2 EXPERIM§NTAE SYSTEM
First the development of the applicator used in these tests is
described. Next the basic experimental system used for these tests is

described.

142

143
8.2.1 MICROWAVE APPLICATOR

The quartz nozzle of the thruster shown in Figure 7.1 limited the
maximum specific impulse that could be obtained. A new design was built
and tested using a metal nozzle and is shown in Figure 8.1. The
numbering system of Figure 7.1 is continued with the changes noted. The
shorting plate (3) is still fixed in position during operation but is
designed so that it is removable, allowing different base plate, nozzle,
flow and discharge chamber configurations to be tested. The discharge
(4), which could be viewed through a brass screened window (5) is formed
inside a quartz discharge chamber (6), in the center of the cavity. A
combination base plate and advanced discharge chamber is showed attached
to the base of the applicator. Gases enter through the base plate and
distributed evenly by a gas flow system that helps to cool the nozzle
and preheat the input gas. The nozzle itself has removable inserts
allowing different high temperature materials to be placed in its
throat.

Figure 8.2 shows the base plate system of the thruster shown in
Figure 8.1 and continues the numbering system of this figure. Gases
enter through the gas input port (16) and distributed evenly by a gas
flow system (17) that helps to cool the nozzle and preheat the input
gas. The gas is injected into the discharge chamber via a circle of
small holes (18) and exits via the nozzle (13) and nozzle insert (15)
after passing through the discharge zone (4). A large vacuum plate (21)
attaches the applicator to the vacuum system. The vacuum plate is
sealed to the discharge chamber by an O-ring (19) and the plate is kept
cool by a water cooling channel (20).

The gas flow pattern is shown by the arrows in Figure 8.2. The

144

Cross Section of a Cavity Applicator
Based Microwave Electrothermal
Thruster with a Metal Nozzle

Legend

(1)-Cavity Walls
CD-awhgsmut
(3)-Bose Piote
(4)-Plasma Discharge
(5)-Viewing Window
(6)-Discharge Chamber
(7)-Coaxial Microwave Input
(8)—input Coupling Probe
(9)-Brass Collars
(13)-Nozzle

(15)-Nozzie Insert

 

 

L__

[—4

_1

E:

 

 

 

 

[1
--—-—-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 8.1

 

 

 

 

:

II:

 

 

 

 

 

 

 

 

A:l

 

 

 

 

 

 

 

 

 

 

.1__
£3 \‘~ 15

Cross Section of 0 Microwave Thruster with a Metal Nozzle

 

 

 

145

Cross Section of a Cavity Applicator
Base Plate

 

Legend
(3)-Fixed Shorting Plate

(4)-Plasma Discharge

(13)-Nozzle

(15)-Nozzie insert A
(16)—Gas input Port 3 a

210
/
(17)-Gas Flow System Z
l?
/
1
\

 

 

 

 

 

(18)-Chamber Vent Holes
(19)-O—Ring

(20)-Cooling Water Channel
(21)-Vacuum Plate

 

 

(-

[@1717

 

 

 

 

 

 

 

 

 

 

00
Till—{-

-------- -Propellant $1» -'
W ~15
+13
-<,{’

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ _____ _ _/ up
I I s
4 2 1‘:
5w
4' 17
i
“\J l
i I<—21
III
i——16

Figure 8.2 Enlargement of Cavity Base Plate

146
propellant passes through the flow system in the base plate of the
applicator, and as shown the input gas flows over the outside of the
nozzle to help keep the it cool. It then enters the discharge chamber
through a ring of small holes at the base of the discharge chamber. The
propellant then flows between the discharge and the chamber wall cooling
the wall and also keeping the discharge from contacting the wall. The
propellant then flows back, partially through the discharge and
partially around it. The heated propellant is then exhausted through
the nozzle which is pointed in an upward direction.

The discharge chamber (4) of the thruster shown in Figure 8.1 was
constructed by epoxying a 25 mm i.d. cylindrical quartz tube onto a
brass base plate. In low power test experiments, with the incident
power limited to 500 Watts, the epoxy would melt limiting operation of
the thruster. The discharge chamber in Figure 8.1 was replaced with the
chamber (4) shown in Figure 8.3. This chamber was shaped like half a
sphere and was of fused quartz construction.

Experiments with the thruster of Figure 8.3 were conducted with
power levels of up to 1.5 kw in helium and nitrogen. However stability
problems with the discharge formation gave very poor performance
results. The discharge would not form in the center of the chamber as
shown in Figure 8.3. Instead it formed near the walls and would move
around the chamber in a random fashion. The discharge chamber
configuration shown in Figure 8.4 was tested because of this problem
with discharge stability. This discharge chamber was also a quartz
hemisphere but it has a 25 mm i.d. cylindrical quartz tube extending
from the top of the hemisphere. A 18 mm i.d. cylindrical quartz tube

was added down the center to help stabilize the discharge. In addition

147

Cross Section of a Cavity Applicator
Based Microwave Electrothermal

Legend

(1)—Cavity Walls

(2)-Sliding Short

(3)-Base Plate

(4)-Plasma Discharge
(5)-Viewing Window
(6)-Discharge Chamber
(7)-Coaxial Microwave input
(8)-input Coupling Probe
(13)-Nome

(15)-Nozzie Insert

 

Thruster

L

P

j

 

 

 

i::l

 

 

r

L .F—1

 

e—l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 8.3 Microwave Thruster with o Spherical Discharge Chamber

 

148

Cross Section of a Cavity Applicator
Based Microwave Electrothermal
Thruster used in Experiments

Legend

(1)-Cavity Walls

(2)-Sliding Short ]
(3)-Base Piate \I/
(4)-Plasma Discharge
(5)-Vlewing Window

(6)-Discharge Chamber
(7)-Coaxial Microwave Input

(8)-input Coupling Probe
(13)-Nozzle
(15)-Nozzie insert : ]

---- ~Propeilant t, I
Flow

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l

 

 

 

 

 

 

 

 

 

 

 

9:1 -
..- -- -""_""—"' \
i | ':
lk

\

 

 

 

 

1
l3

Hoe—1

 

 

 

 

 

 

 

 

 

 

 

 

 

:1":
V I:: q
«1
[.___l. 4h—
I‘ J
W Ls
5—-o
I 3
I I
, .\\\\\‘

 

 

 

 

 

 

15

Figure 8.4 Microwave Thruster used for the Experiments in this Chapter

149

to the gas flow path through the base plate, there was gas flow down the
center of the second quartz tube. This configuration produced excellent
results using nitrogen but still had some stability problems in helium.

Results for both gases are presented in section 8.3.

8.2.2 EXPERIMENTAL APPARATUS

The basic microwave power and measurement system used for these
experiments was described in Chapter 4. The thruster measurements were
performed with a continuous wave 2.45 GHz variable power (from 0 to
2500 Watts) microwave system. Experimental measurements were performed
using nitrogen and helium gas as propellant, with flow rates of up to
31.3x10"6 kg/s.

The base of the energy absorption chamber was attached to a vacuum
chamber that included a heat exchanger to cool the exhausted propellant.
The pressure and gas flow rates were measured using the measurement
system of Chapter 4. The measurements were performed with the
applicator in a vertical position as shown in Figure 8.5. As shown in
this figure the nozzle of the thruster was above the discharge. As
mentioned in Chapter 7 for the experiments by S Nakanishi, it was felt
that this orientation helped the operation of the thruster.

Measurement of hot and cold discharge chamber pressure for
constant propellant flow rates were performed for different levels of
input power, different propellant flow rates and variations in discharge
chamber configuration. Using the rational of the indirect thrust
measurement described in section 4.4.3, energy efficiency, specific
impulse, thrust and thrust to power ratio were calculated using

equations 2.1-2.5 and 4.6-4.9.

150

EXperimentaI Setup of
Microwave Electrothermal
Thruster used in Experiments

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_ _‘—>2" Pyrex Glass
5” Pyrex Glass Vacuum Line
Vacuum Chamber , t0 VOCUUI’T‘ Pump
Cooling
Water
Heat 0'”
Exchanger
>
Cooling
A Water
In

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

U P I
Thruster
(Horizontal Axis . From
Not To Scale) Flgure 8.4

 

 

 

 

 

 

 

 

 

 

Figure 8.5 Experimental Setup of Microwave Thruster

151

8.3 EXPERIMENTAL RESULTS

Experimental runs were performed by first establishing a desired
propellant flow rate and holding this flow rate constant throughout the
entire run. The cold discharge chamber pressure, pc, was measured. The
discharge was started and the input microwave power was adjusted, and
' the applicator input power, Pt, and the steady state "hot" discharge
pressure pH were measured. After each experimental run, the discharge
chamber was allowed to cool back to "cold" conditions as a check against
significant nozzle erosion. As discussed in Chapter 5, microwave
coupling efficiency is very high, thus Pt is approximately equal to the
microwave power absorbed by the discharge, Pa' and thus the energy
efficiency, the thrust to power ratio and the specific impulse can be
calculated. Figures 8.6-8.8 summarize the experimental results where
the legend in Table 8.1 details the different experimental
configurations and operating points.

Figures 8.9-8.11 show these results compared to earlier results
with quartz nozzles. The nozzle used in this test was very small and
designed mainly for use with helium. This limited the maximum flow rate
that nitrogen could be tested at to 31.1x10"6 kg/s since higher flow
rates produced a discharge chamber pressure that was to great. As can
be seen in the figures the results for nitrogen are very good if the low
gas flow is taken into account. Results for helium are not as good as
earlier result with the quartz nozzles. This is due to the instability
of the helium discharge inside the discharge chamber. Nozzle erosion or

melting was not observed during the 31 hr. of experimental Operation.

152

Table 8.1 Operating Conditions for Experiments Presented, in Chapter 8

Operating Conditions for
Experiments with the
Microwave Thruster of Figure 8.4

 

F low Discharge Cavity Nozzle Dual
1' Point Rate Pressure Mode Size Flow LS. LP

(kg/sX1D'6) (Torr) 1:11 cm

1. 3 10.4 212-247 TM011 0.351111 yes 6.85 1.1
2. 0 20.8 645-690 TM011 0.351111 yes 7.25 1.8
3. O 31.3 1115-1147 TMou 0.351111 yes 7.25 1.8
4. 3 8.9 382-465 TM 011 0.351111 yes 7.0 1.85
6. O 17.8 1040-1080 TMDll 0.351111 yes 6.9 1.92

 

Solid Points — Nitrogen Propellant
Open Points -— Helium Propellant

%

Energy Efficiency

153

Energy Efficiency Versus
Input Power
for Operating Conditions

 

 

of Table 8.1
50
40 —
30 ..q R0
O\Ooo ‘\.
20 — 8\ \.
88 _
10 _ e e.
X—x—x
0 I I I I

 

 

0 800 400 600 800

Power — Watts

Figure 8.6 input Power Versus Specific Impulse

1000

En erg y Efficiency %

154

Energy Efficiency Versus
Specific Impulse
for Operating Conditions

 

 

of Table 8.1
50
4am
30 -‘ Q0
A 0%
201 a:
\%
10 - .10
“I
0 I I I T I I

 

 

0 100 200 300 400 500 600

Speciﬁc Impulse - Seconds

Figure 8.7 Energy Efficiency Versus Speciﬁc Impulse

700

155

Specific Impulse Versus
Thrust to Power Ratio
for Operating Conditions

 

 

 

 

of Table 8.1
103
.1

in
t, ..
C
8 .4
1% at

O
I '— QDCNo
o ’38

‘O
Q s
E. ’.
O
9;: _.
8
1%- K,
102 I I I I I I I

0.05 0.1
Thrust to Power Ratio

Newtons/kWatts

Figure 8.8 Speciﬁc impulse Versus Thrust to Power Ratio

0.5

Energy Efficiency %

156

En ergy Efficien cy Versus
Input Power

for Operating Conditions
of Tables 7.1, 7.2 and 8.1

 

 

 

 

50
Z\<
40 - ox
v O
A R
30 — \&X¢\
0%.“ . A
\ \‘\‘
80 — EI\ ‘Q\ ...—*“ﬂ
x 6) N Q—F‘L—‘P'
10 — ":ON H—e—e
XXx
0 I I I I I

0 400 800 1200 1600 2000

Power — Watts

Figure 8.9 Energy Efficiency Versus Input Power Comparison

Energy Efficiency 73

50
<1
<>\<I
40- \ 9,
Q,
of A
30- O a
‘0 e O
59 ‘7 Q” A
80— *1} D\
\
Q§§ a
10 _. '8
e
h
0 r I I I I I

157

Energy Efficiency Versus
Specific Impulse

for Operation Conditions

of Tables 7.1, 7.2 and 8.1

 

 

 

 

0 100 200 300 400 500 600 700

Speciﬁc Impulse - Seconds

Figure 8.10 Energy Efficiency Versus Speciﬁc Impulse Comparison

158

Specific Impulse Versus
Thrust to Power Ratio

for Operation Conditions
of Tables 7.1, 7.2 and 8.1

...s
O
(a)

 

Speciﬁc Impulse — Seconds
a?“

 

 

 

102 I In I I I I I
0.05 0.1 0.5
Thrust to Power Ratio
Newtons/kWatts

Figure 8.11 Speciﬁc Impulse Versus Thrust to Power Ratio Comparison

159
8.4 DISCUSSION OF RESULTS

The thruster had excellent results in nitrogen at very low flow
rates compared to earlier experiments with nitrogen at higher flow rates
using quartz nozzles. Further, unlike earlier experiments with nitrogen
in a TM011 mode using a quartz nozzle, there was no observed nozzle
melting. The use of the metal nozzle eliminated all problems with
nozzle erosion for these initial experiments.

The results with helium were not so good. While the nozzle
melting problem was solved by the use of metal nozzles, the helium
discharges could not be formed in the optimal position for the best
energy transfer to the propellant gas. Further experiments are needed
to study this and to design a new internal flow system and discharge
chamber that can control the position of the discharge for optimum

energy transfer to the propellant.

CHAPTER IX

CONCLUSIONS

9.1 SUMMARY OF RESULTS

Initial experiments demonstrated the ability to generate microwave
discharges at atmospheric pressure in He, N2 and 02 gases. These
discharges were stable from 40 to 1000 Torr in pressure with a input
microwave power of 500 Watts. Coupling efficiency of microwave power
into the discharge was measured to be in excess of 95% over a pressure
range of 40 to 760 Torr. As pressure increased, the discharges were
observed to contract from the chamber walls forming intense small volume
discharges with high power densities. Different diameter discharge
chambers were tested to check their effect on discharge properties.
Chamber wall-discharge interactions were found to have a great effect on
discharge properties. The experiments demonstrated the basic components
necessary to produce a working microwave electrothermal thruster.

The experiments have demonstrated a working microwave
electrothermal thruster thus proving that the concept works. Two
different microwave design concepts were employed to produce a working
electrothermal thruster. The first of these devices utilized a coaxial
microwave discharge to heat the propellant. This device operated at a
microwave frequency of 2.45 GHz in a TEM mode and was tested in a power
range from 200 to 600 Watts. Experimental measurement of thrust,
specific impulse and energy efficiency were taken for nitrogen gas with

flow rates up to 10.6xl0‘6 kg/s. Measured thruster energy efficiency

160

161
varied between 308 to 60%.

The second of these devices employed a cylindrical cavity
microwave discharge to heat the propellant. This device operated at a
microwave frequency of 2.45 GHz in a TM012 cylindrical cavity mode in
power ranges from 1400 to 2000 W. Experimental measurement of thrust,
specific impulse and energy efficiency were taken for nitrogen gas with
flow rates of up to 146x10'6 kg/s resulting in measured thruster energy
efficiencies between 10 to 258 and specific impulse of up to 280 sec.
Experiments with helium gas operating in the TMO11 mode resulted in
energy efficiencies of up to 508 and specific impulse of 200-600 sec.

These measured energy efficiencies and specific impulses compared
favorably with other electrothermal thrusters such as the resistojet and
arcjet. However, both of these devices were designed to demonstrate the
basic concept of a microwave electrothermal thruster and were,
therefore, kept simple in design and were not optimized. For example,
the quartz nozzle used in all of these experiments limited the maximum
specific impulse due to melting failure of the quartz. Forced air
cooling of the discharge chamber was also employed during all
experiments resulting in reduced efficiencies due to increased energy
losses.

A new cavity applicator configuration was designed and tested.

This new design incorporated a metal nozzle with regenerative cooling in
a removable base plate. Experiments with this new thruster design
produced very good results in nitrogen although discharge stability
problems limited results in helium. The stainless steel nozzle showed

no signs of erosion or melting. If the problems with discharge

162

stability can be solved, it is expected the performance will improve

over the results of earlier designs.

9.2 CONCLUSIONS

These experiments demonstrated how to generate a high pressure
microwave discharge under flow and no flow conditions without melting
the enclosing walls. Experiments with no gas flow demonstrated that the
coupling of microwave power into the discharge is a very efficient
process. Plasma volume was measured during these experiments and the
discharge power density was determined versus pressure for He, N2 and
02.

This experimental work demonstrated that the microwave
electrothermal concept is technically feasible and the measured
performance compares favorably with other electrothermal propulsion
concepts. However, many questions remain about its use in a real space

application, but no major problem were discovered during this research.

9.3 RECOMMENDATIONS

In experiments in this thesis measurements with hydrogen as a
propellant were never tried for safety reasons. Most electrothermal
thrusters are tested in hydrogen since it has the highest cold specific
impulse of any gas. The experiments with the cylindrical cavity and
coaxial applicators presented in this thesis, should be repeated using
hydrogen gas. The values of energy efficiency and specific impulse can
be then be compared to other electrothermal thrusters using hydrogen.
Further, these experiments will have the benefit of demonstrating and

improving the understanding of hydrogen gas discharges at microwave

164

works and would provide some direction in the development of new
thrusters. If developed well enough, a working model of the thruster
could be developed and the maximum performance that could be obtained
could be calculated.

To help develop these models measurements of fundamental microwave
discharge properties need to be conducted. Spectroscopic measurements
would give the values of electron temperature and gas temperature inside
the discharge. Identification of the constituent species present in the
discharge would also be of value along with their distribution in the
discharge. Energy balance measurements would serve would serve to
develop a understanding of energy paths that exist in the discharge.

The total microwave thruster system needs to be studied. This
includes not just the microwave electrothermal thruster, but all the
other system components that are required for it to operate a thruster
such as the microwave supply system and the propellant supply system.
Factors such as total system cost, weight, lifetime, etc. need to be
studied. For any real space applications these factors need to be
compared to those of other thruster concepts to get a good understanding

of which concept is best.

APPENDIX A

APPENDIX A

MICROWAVE FREQUENCY COMBUSTION FLAME INTERACTION

A.l INTRODUCTION

A review of previous experiments studying the interaction between
a high frequency electromagnetic field and a combustion flame is
presented in section A.2 of this appendix. Experiments studying the
interaction between a combustion flame and the electromagnetic field of
a microwave cavity operated in a single mode, are presented in section
A.3.

Combustion flames are of great importance to world energy
production. Almost all transportation and energy production in the
world derives its energy from combustion flames. In chemical
production, combustion flames directly produce chemicals or produce the
energy required for their production. In homes combustion flames
provide energy for heating and cooking, and in industry combustion
flames are used in processes such as welding, heat treating, etc..

Because combustion flames are used so extensively, any method to
improve their performance is desirable. It is possible that by adding a
high frequency electromagnetic field to a combustion flame, its
operation could be improved in a beneficial manner. If the efficiency
could be increased, the pollution be reduced or other useful benefits be
found from these interactions, this research would prove to be of great

value.

165

166

As an example of one potential benefit, the use of microwaves to
assist combustion in an internal combustion engine is detailed by Ward82
and, Ward and Wu83. In order to reduce pollution in an internal
combustion engine a leaner fuel/air mixture can be used. However, this
results in a lower flame speed which results in lower efficiency. It is
suggested in the first paper that microwave energy could be used to
increase the flame speed, thereby increasing efficiency. The second
paper presents a theoretical model that indicates this may be possible.
One factor in favor of this conclusion is the greater number of
electrons present in a hydrocarbon flame a microwave field could
interact with. An electron density of 1011 electrons/cm3 is stated in
the paper for lean hydrocarbon flames while for H2 and CD a figure of
106 electrons/cm3 is given.

The exact definition of a flame is rather vague, but for this
appendix a flame can be considered a highly exothermic reaction between
gases. It is also useful to define some other terms.8‘v85

Diffusion flames: In a diffusion flame the flame occurs at the
boundary between the two gases. The rate and manner in which they
diffuse together controls the direction and rate of propagation of the
flame front. This is shown in Figure A.1a.

Premixed flames: In a premixed flame the two gases are mixed
before being burned. A term called flame velocity can be defined as the
rate the flames moves through the premixed gases. This is shown in
Figure A.1b.

Flame Plasma: The flame plasma is the plasma or region of peak
ion/electron interaction that exists in the reaction zone of a flame.

As shown in Figure A.2, a flame can be divided into several zones.

167

Diffusion and Premixed Flames

 

 

 

 

 

Direction
°f C A
Diffusion ' O S '
i i
I i
i l
\\ : [I \\ ll \\ g [I
\‘V’ \‘ll/ \‘V’
4. 4r 4.
I'\ I \ I \
l/ : \\ ’1 \\ [I i \\ A
.__.. I I
i I
i i

 

 

Gas B

 

 

 

 

 

 

Flame
(0) Diffusion Flame
I
‘-—-Flome Velocity
V
Flame

Gas A and Gas B

(b) Premixed Flame

Figure A.l Diffusion and Premixed Flames

168

Flame Zones

 

 

 

Temperature
I I
I
I I
I g\_ Ignition
I I Point
I I
l I
I I
I I
I I
I I
l I
I I
I I
I I
I I
l I
I
I I
I I
I I
I I
I I
I I
I Preheat I Reaction
! Zone 1 Zone

Figure A.2 Flame Zones

169

In the preheat zone heat from the chemical reaction heats the gases
entering into the region in which the chemical reaction is taking place.
The ignition point is the point where the combustion starts. The
reaction zone is the region in which most of the combustion takes place.
Many radicals, ions and electrons are present here.
For example, in a hydrogen-oxygen flame above 600 C some of the

radical reactions present in the reaction zone are:86

H + 0H + M > H20 + M

H+H+M<>H2+M

0 + 0 + M <> 02 + M

A. 2 93311415111“ REVIEW

Figure A.3 shows an experimental set up used by Jaggers and Van
Engel.87 As shown it consisted of a vertical convection free tube which
was evacuated, then filled with premixed gases. A flame was ignited at
the top of the column and allowed to propagate down the column. Two
diodes placed a known distance apart were used to measure the time it
takes for the flame to travel between the diodes. The flame velocity
was then calculated from this data. A transverse electric field could
be set up by applying a voltage between the two 90 cm long aluminum rods
fitted to the inside wall of the tube. These remained in position
throughout the experiment to retain the uniformity whether or not the
electric field was used.

Initially town gas (55% H2, 20% CO, 20* methane and 5% N2) was
used as fuel and mixed with air. No measurable change in flame velocity

was observed when an electric field of up to 1kV/cm in a do or 50 Hz

170

Experimental Setup
of daggers and Van Engel

 

5cm

 

II

 

 

 

 

 

 

 

.L___s

 

 

 

k Photocell

 

 

 

120 cm /\ A
I
i I/

 

Intern ol
II Electrodes

 

 

(4 Photocell

 

 

.L___J

 

 

(Redrawn from
Reference #87)

Figure A.3 Experimental Setup of Joggers and Von Engel

171

electric field was applied. Several reasons were given for the lack
increase in flame speed. Two of these were non-uniform mixing and lack
of hydrocarbons present in the mixture.

For hydrocarbon flames the results were better. For a 10%
methane-air mixture in an electric field of 440 V/cm at 5 MHz, a 10 cm/s
increase in flame velocity was observed over the no field case. A
definite increase was observed for methane-air mixtures of 7.5% to
11.5%. For a 5% ethylene-air mixture in a electric field of 440 V/cm at
5 MHz an increase of 50 cm/s was observed over the no field case. For
ethylene/air mixtures between 4.28 and 5.42 a large increase in flame
speed was observed and for mixtures up to 6.58 a smaller increase was
still observed.

Figure A.4 shows a second experimental setup that was tested using
a floating flame. A long 5 cm diameter tube served as the flame
apparatus. A premixed gas flow was passed up through the Pyrex tube.
By adjusting the velocity of the gas the flame could be held stationary
in the glass tube.

When the electric field was added the flame would move down an
amount 2. For both methane and ethylene there was a measurable change
in z in the presence of an electric field. This change reached a
maximum of 1 mm for methane with an electric field strength of
800 V/cm rms at 7 MHz.

daggers and Von Engel felt that the main effect the electric field
had on the flame was to change the properties of the electrons in the
flame front. Assuming all the current flows through the flame reaction
zone, the electron density can be determined by the current flow

assuming the flame front is uniform. This gives a electron density of

172

Floating Flame Apparatus
of Joggers and Van Engel

 

 

 

 

 

 

 

 

 

 

 

 

I; 5 cm =
I) I ——————————————— — ————————————— I
‘-—- Metal Gauze' g
Electrodes
Flame
T V
Z
10 cm __I_
Premixed
Gases
._____ Cotton
Wool
II

 

 

 

 

 

 

*2 cm—-'*

(Redrawn from |\
Reference #87) __ __ Pyrex

 

 

Figure A.4 Floating Flame Apparatus of Joggers and Van Engel

173

ion/cm3 for ethylene and 5x1010/cm3 for methane.

Jaggers and Von Engel give a value of 10,000 Watts for the energy
of the chemical reaction. Since the electrical field generated a
current of less than 0.5 mA the energy coupled by the field into the
electrons is under 1 Watt. If this were dissipated in the ion or
neutral gas this would produce an increase in temperature of only 4 K,
but if it were dissipated in the electrons it would produce an increase
of 4000 K in the electron gas temperature. The authors felt this
increase in temperature is responsible for the observe experimental
results.

Figure A.5 shows an experimental setup used by Tewari and
Wilson.88 The flame chamber is first evacuated then filled with the
desired air/fuel mixture. A laser then ignites the mixture and a high
speed camera takes pictures that are used to determine the flame speed.
Two electrodes placed in the top and bottom of the flame chamber
introduce RF into the flame.

The authors tested methane/air mixtures, methane/oxygen/argon
mixtures and hydrogen/air mixtures. Using an electric field of
1.67 kV rms at 6MHz and a electrode separation of 2.2 cm, results for
methane/air give an increase of as much as 100% for flame velocity under
some conditions. The methane/oxygen/air mixture had less RF influence
than the methane/air mixture. This was contrary to what the authors
thought would happen. They felt that neutral argon gas would absorb
less energy from the electrons than neutral nitrogen gas so the flame
velocity would be larger. For a 10% hydrogen/air mixture there was a
very slight increase in flame velocity.

Figure A.6 shows an experimental setup used by MacLatchy, Clements

174

Experimental Setup
of Tewari and Wilson

(Light from Flash Lamp)

/- Window
Window /

\ Window
(Light to Camera)

Top View Lens

.—

 

 

Electrodes

\

\l
"TTTFTTTTTTTTTT"

--rF---‘-;:::

/

Ignition
Ch amber—7

 

 

 

T

I...

____I_______L
\
____FC__-_-__

Electrode

Window —\

I I

. . ‘.::::‘$::::: ::::::::; (Loser
IgnItIon _> I“ _________ . _________ Beam)

Chamber \ _____ \
T \
l | Window

(Redrawn from Electrode
Reference #88)

 

 

 

 

Lens

Side View

Figure A.5 Experimental Setup of Tewari and Wilson

175

Experimental Setup of
MacLatchy, Clements and Smy

(Rembwn ﬁom
Reference #89)

 

Langmuir Chimney
Probe
I /— Flame

 

 

 

 

 

 

 

 

 

 

 

 

Microwave Oven

Figure A.6 Experimental Setup of MacLatchy, Clements and Smy

176

and Smy.39 The experiment was performed in a commercially available
microwave oven. The oven operated at 2.45 GHz and generated 500 Watts,
although the magnetron operated on only half cycle of the electrical
alternating current supply. A Meker style burner made entirely of pyrex
glass was used and propane was used as a fuel. The burner was placed in
an experimentally determined point of high electric field. The flame
was started the top of a chimney attached to the burner, outside the
microwave oven. Then the gas flow was turned off and the flame would
propagate down the chimney, into the microwave oven and the flame speed
could be measured.

A Langmuir probe was used to measure electron temperature and
ionization density in the flame front. Results of the probe gives an
increase in electron temperature of 4000 K in the presence of microwaves
compared to 300 K for no microwaves. However, no increase in flame
speed was observed.

Figure A.7 shows the experimental setup of Clements, Smith and
Smy.90 As shown in the figure the experimental set up consisted of a
magnetron tube connected directly to a cylindrical cavity. The cavity
was tuned with a quartz rod to a resonant frequency of 2.448 GHz when
loaded with the flame and was excited in a TM01O mode when empty with a
power level of 1800V/cm rms at the center. A 22 mm i.d. cylindrical
quartz tube ran axially through the cavity. The premixed gases were fed
up through the tube and the flame was started at the top. The gas flow
was shut off and the flame was allowed to propagate down the tube
through the cavity. Two photo diodes at the ends of the tube were used
to measure flame speed.

Results were given for propane/air, ethylene/air and acetylene/air

177

Experimental Setup
of Clements, Smith and Smy

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Quartz Tube
Quartz Tuning Rod
_\\\\\\ 5 ’//,//— RF Choke
Magnetron - ﬂ ' Photo Diode
’M/
I 0
II I I I
[—1 I
* J I I I I
IIE3] g t:
c r , 5' RFlOut
CUP ”19 I I (To Crystal
WavegUIde ’I Detector)
Coupling Hole - g I

 

 

 

/

""”V""
O
————-—- -_——4 —-._———_—

RF Choke-”/////' \\\L_ .
Photo DIode

Bunsen Burner

 

 

 

 

(Redrawn from Reference #90)

 

 

 

 

Figure A.7 Experimental Setup of Clements, Smith and Smy

178

mixtures. The flame speed was found to be stronglydependent on the
temperature of the quartz tube so the temperature of the quartz tube was
held constant using forced air cooling. The experiment showed small but
significant enhancement of flame speed under conditions with no gas
breakdown by the microwave field. This ranged from a maximum increase
of 203 to a decrease of 58. With a gas breakdown, increases of up to
403 were observed.

The experimental setup shown in Figure A.8 was used in experiments
by ivlard.91'92 As shown in the figure the experiment used a cylindrical
bomb. This device is basically a cylinder filled with premixed gases,
then the mixture is ignited and the mixture burns very rapidly inside
the cylinder. In this experiment the cylinder was designed to operate
as a TM010 mode cavity at 2.45 GHz. The cavity was evacuated and then
filled with a propane/air mixture. The mixture was then ignited and
measurements were taken using a RF probe, a Langmuir probe and a
pressure transducer. A increase of up to 2 times is reported for the

flame speed with the addition of the microwave field.

A.3 MICROWAVE-COMBUSTION FLAME RESEARCH

Initial flame experiments were conducted using the setup shown in
Figure A.9. The microwave cavity is described in section 3.4 of this
thesis. A candle was positioned so that its flame was at the center of
a TM012 cavity mode. Input power was supplied be a 100 Watt continuous
wave microwave source. The cavity was set off resonance and the
microwave power would be turned on. The candle was lit and placed in
the center of the cavity. Then the cavity would slowly be tuned to

resonance. As the cavity was adjusted to resonance the flame would

179

Experimental Setup
of Ward

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Spark Langmuir
Plug—q __Prabes
Pressure
Transducer—1
III II
JU— L—FTI—UL —..——
RF In RF Out
Cylindrical
381: Combustion A
Bomb
Air/Fuel
Inlet and 1
Exhaust —-—'—
Valve
!: 38 mm =

 

A = 6.3 to 25.4 mm

(Redrawn from Reference #91)

Figure A.8 Experimental Setup of Ward

180

Cross Section of
Flame Experiment

— _

 

 

 

 

 

 

 

 

Legend
(1)—Cavity Walls
(2)—Sliding Short :1
(3)—Base Plate

(5)—Viewing Window
(7)—Microwave Input

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(8)—Coupling Probe C. 1 II J
2
Ls
LP.
7 ‘_ T f1
L._ )+8 5”
C ﬁn I II
I I
I] l
3
/I\
Cande

Figure A.9 Cross Section of Flame Experiment

181

suddenly increase in length until it was about 10 cm long. It would
stay like this for about 15 second and then go out. The combustion
products of the flame were found to be contaminating the walls of the
cavity so the experiments were discontinued. The cause of the observed
effects was difficult to ascertain. It was possible that the microwaves
were directly coupling into the flame. However, it is also possible
that the microwaves were being coupled into the candle wax and heating
it, thereby providing more fuel to the flame.

A advanced experimental setup was constructed and is shown in
Figure A.10. The same microwave cavity was used but a quartz tube was
added to keep the flame combustion products off the cavity. The flame
source was a Bunsen burner using propane. The cavity was tuned to a
TM012 mode (shown in Figure 3.8) and a piece of screen was placed across
the bottom opening to increase the field strength in the flame. A
microwave sweep system, shown in Figure A.11, was used for the input
power to the cavity. The cavity was tuned for a center frequency of
2.45 GHz with a bandwidth of 6 Mhz and a input power level of 25 Watts
from a traveling wave tube.

Figure A.12 shows the sweep results for propane input pressure of
6 lbs/in2 to the bunsen burner and Figure A.13 shows the results for 3
input pressure of 7 lbs/inz. With the addition of a flame to the
cavity, a frequency shift of 0.2 Mhz and a slight drop in Q is observed.
This indicates that the presence of the flame is affecting the microwave
field inside the cavity, thus changing the cavity resonance. There are
two possible reasons for this change in field pattern. First, the
microwave field may be coupling into the flame, affecting the field

pattern. Second, the hot flame may be affecting the dielectric constant

182

Cross Section of a
Flame Experiment
using 0 Bunsen Burner

£5 :0:

Quartz
Tube

I..|‘ ‘

 

 

 

 

 

Legend I
(1)—Cavity Walls

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(2)—Sliding Short 1
(3)—Base Plate -—-——
(57‘V19WIng Window Flame
(7)—Microwave Input 1 a J r——-4 ...—J
(8)—Coupﬁng Probe d I I b
2 Ls
LP
7 : c=_—-—-—" 9 ﬂ]
LT:—. :3: 3‘f3 ‘5_H.
II I I II fJ

 

 

 

l
3
Bunsen Burner—a L

Screen

 

 

 

 

 

 

 

 

 

 

Figure A.lO Cross Section of 0 Flame Experiment with a Bunsen Burner

183

EXperimentaI Microwave
Sweep System

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

incident a
Power Meter MI
Sweep crowave
Generator J Applicator
» Attenuator ,
L83?” I\ RF Out
Traveiin
Wave 9U Directional J I
THIN I [Coupler I K
CIrcuIator
Matched
Load
I Directional I
I f CoUpIer I
. , Crystal
X—AX'S Detector
Input Y—Axis
/— Input
Matched Load
1b
X-Y Plotter

 

 

 

 

 

 

 

Figure A.11 Experimental Microwave Sweep System

184

Experimental Results

Reflected Power Versus Frequency

        

I ‘ I 'III . ' ' " ‘ ' IIIIII 'l IHIIH' IIIIIII I IIIIII. IIIIIIIIIIIIIIIIIIIIII‘IIIIIIIII IIIIIIIIII
...... ""”IH'IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHillIIIlIIiIII'IIIIIIIII'IIII I: RWHMHHHIHIIH IIIIIIIIIIIIIIIIIIIIIIIII‘IIIIIIIIIIIIIIIIIII II'II I’HHIIII Ia
II “Illlirﬂrm ”1"“111 IIIIlIIIlIIIIIIIlIIIIIlHIII {lllllllllmgllliilllillllllll IIIIIIIIIIIIIIIIIIII HIIIIIIIIIIIIIIIIIIIIIIIIIIIHIIIIIIIIINI‘IIIL‘IIHIIIL

m IIIIIIIIIIII IIIIIIIl IliIlIIIIIH'" IIIIIIIII IIIHIIIIIIIIIIIIIIIIIIIIIIEIHH'II Illilllﬁ%mlll W “W HiIIIIIIIIlIIIIIIIIHIHIIIIIII IIIIIIIIIIIII‘IIIIIIIIIII' IIIIIIII IIIIIIIIIIIIIIIIIIIIIIII‘IIIIIIIIII IIIII
IIIH IIIIIlIlIIllIIIIIIIIlll'IIIII‘IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII.IIIIIII"laI."IIIH. IIIIIIHIIHII IIIIIIII llIl‘illHllillIHIIII llllIIlIIIIHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII.I III" IIIIIIIIIIIIIII IIIIIIIIIIIIIIIII IHIIII‘IIIIaI
EHIIIHIIIHI IIIIIIIIII IIIIIIIIIIIIIII' lllIIIIIIIHIIIIIIII IIIIIIIIII IIIIIIIII Will 11an IIIIIIIIIIIIIIHIIIIIIIH IIIIIIIH IIIIIIIIIIIIIIIIIHIIIIIIIIIIIIII‘IIIIIIII'IIIIIIIIIILIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII‘II‘“ .II
III'IIIIIII IIIIIIiililI lilIllllIIIIII IIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIllIII'IIIIIIIIlI "IHI IIIIIIIIIIIIIII IIIIIIIIIIHIIIIIIIIIIIHI IHI!!! ﬁlﬂilllllliillllll IIII'IIIIIIIIIIIIIIIII‘IIIIII’IIIIIIII‘‘I IIIII’ IIII llll‘i. "a’lll IIIIII IIIIII" .IIIHIIIII III

I“ I

 

IllIII Illlll IIIIIIllIIIIIIIIlIlIIllIlllliIIIIIHIIIIIIII. II'IIIIIIIIIIIIH III 'IIIIIIIIIIIIII 'IIIII."III I II"IIII 'III IIIIIIIIII IIIIIIIIIIIHI HIIIIII IIIIIIIIIIIIIIlIIlIlIiIIIIIIIIIIIHIIIII IIIIIIIIII IIIIIIIIIIIIIIII IIIIIIIIIIIIII IIIIIHIIIIIII: .‘IIIII Illlllll .III 'III (Illll
I‘ll III IIIIIIIIIIIIIIII‘IIIIIIIIIIIIIIIIIIIIHIIIIIIIIIIIIIlllIIIIIIIIIIIII‘IIIIIIIIIIIIII'IHI "III IIIIIIIIII IHIHMHIIIHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHIII IlIIIIIIIII IIIHII IIIIIIIIIIIIIHIIIIIIIIII HI aIIlIIIiIIa‘IIIIIIIIII IIIl IIIHIIaIIIII” IIIlaIIIII Illlll I'llll
IIHIllIIIIIII'HIIHIHIHIIHIHIIHIHIIIIHIIIIHIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIlIIIIII IIIIIIIII lHIIE'IIlIIIIIIIIIIIIIHH IIIIIIII IIIIIIIIII IIIIl IIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIHIIIHI IllIIIIIIIIII‘IIIIIIIIIIIIIIIIIIIII |!IIIIll ",II‘ IlllllIIIlIIlIIII IIIIIIIIIII
LIIIIIIIIIIIII‘IIIIIIIIIIIIIIIIIIlIllll IHIHIHIIIII llIIIIIIIII IIIIIIIIHIIIIIIIIIIIIIIIIIIIIIIIIIIII'HII lIIIIH‘IIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIHIIIIIIIIIIIIiIlIIIIIIIIIIIIIIIHIIIIIIIIIEHIHIIIIII‘IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII‘I‘ .IIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIII
IIIIIIIII IlllIlIl IHIIIIIIHIIIIIIIIIIIIIIIIHIIIIIIIIm‘ IIIllIIlIl IIIIIIIIIIIIIIIII III “III lIIIII IlIIII lilIIIIIIIH 'liIlII IIIIIIIII IHIIIIIIIIIIII IllIIHIII IIIIIIIIIIIIIIIIIII IIIIIIII IIIIalII IIIIIII IlIIIHIIIIlHIIlIIIIIIIIIIII‘lIIIlIIIIIaIIIIIIIIIII IIIl'" .IaIIIl'IIIIllIIIIIIIIiIIII'III‘IIIIIIIII
I'll‘ :lllllllll IIIllIIIIIIIIIl IIIIlIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIII IIII IIIIIIIIII IIIIIIIIlIlInIIIII‘aIIIIIIIII lIII I1 I'l'IIIIIIIlIIIlIIIlII HIIIIIIllI IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIII'IIIII‘IIIIIIIII 'II [III iIIIIIIIII l iIII IalII.‘ ‘II“a‘II II IIIIIIII IHIIIIIIIi III‘III
.I‘IIII‘IIII‘IIIIIIIIIIII'IIIIIIIIIIIIHIIII IIIIIIIIIIII IIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 'II IIIIIII IHIIIIIIIIIIIIIIIIIII‘ IIIHIII IIIIIIII'IIIIIIIIIIIIIH IIIIIII II'II hIIIll‘ ' ' II

;::E

   
 
  
  
 

 
 
 
 
 
 
 
 
  
 
 
  
 
 
 
 
 
  
 
 
 
 
 
  
 

 

IIII‘ I ‘IIIIIIIIIIIII IIIIIII I Illlllélwllll IIIIIIIIIIIIIIIIIIIIIIIIII'IIIIIIIIIIIIIIIIIIIIIIII HIIIIIII I'IIII III). IIIINIIIIH IIIIIIIIIIIIIIIIIIIII ' '

III‘I IIIIIIIIIIIIIIIIIIIIIIIIIIM IIIIIIIII IIIIIIIIIIIIIIIIIIIII IllllilI IIIIIIII'IIIHIIIIIIIIIII IIIIIIII IIlIIIiI IIIIIIIIIIIIIII FICil'Tl

FIII ‘IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHIIIIIIIIIIIIIII'IIIIIIIIIIIII lIIIIII IIIIIIIIIIIIIII.» I."‘I‘I II '

IIII llIIlIlllllIlIIIlIlIIliIlIIIII llllIIIiIl IHIIWIIHIHIIIIIIIIIIIIIIIIIIIIIIIIIHIIIIIIIIII IllIIIIiIIIIIIIII IIIIIII'IIIIIIIIIII Ii ||IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHIIIIIIIIIIIHIIIHIIIIIIIIlIIIIIIlII'II' II'.IIIIIIII‘H‘IIIIIllllllIIIiIIII‘IIIIIIIlIIlIII'IIIII Illlilllllll
aIII R aIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHHIIIIIIIIIIIIIIIIIlIIIIlIIHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIHIIIIIIII IIIIIIII; IIIIIiIaII. IIIII'IIIIIIIIIIIIIIIIHIII IIHHIIIIIHIIIHIIHIHIIHIII.“ Ill' .Il'l aIIIIIIIIIHIl‘lIIIIIIIIII IIIIII'IIIIIIIIIIIIIIIIrIIIllIllIIIIII'
IIII IIII'IIIIIIIIIIIIlla'IlIliIIIIIIIIIIIIIIIIIIIIIIIIIIZIIIIIIIII'IIIIHIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIImIIIHIIIIII IIIIHIIIIm IIIIIIIHIIIHIIIIIIMIIHIIIIIIIIIIIIIIIIIHII‘HI‘ Il‘IIIII' 'III‘aII‘IIIIIIIIIIIIII‘IIII‘IIIIIIIIIIIIIIIIIIIII
IIII IIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII‘ HIM“ InﬁlllllllllllllllllﬂllIIIIIiIIIIIIIIIIIUlIIiIIIIII IlIlIIIlI II I Hill I‘IIII. m’IIIIIIIIH IIIH“. I113 HIIIIIIIII IIIIl‘Il
Illl IIIIIIIIIIIIIIIIIIIIIIIIIIIHIHIIIIIII HINIHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII’IIIIIHIIIIHIIIIIHMHIIIIIIIIIIIHI'IIIIIIIIIIHII IIIH‘IIIIIIIII Willdl'zlllllll HIIIHIHIIHI IIIEHI'.
IIII IIIIIIII‘IIIIIIIIIII IIIIIIIIIIIIIIIIIIHIIHIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIH‘ iII'IHlIIIIIIlIHlIIdHII IIIIIIIHI HIIIIIIIHIIIIIIIII IIIHIIIIIHIIIIIIIIHIII IIIHIIIIIIIIIIIIIIIIII IIIIIIIIIIII'IIII'I IHIII IIII'IIIIIIIIIIH I III. IIIIIIIII Ill
I'll llll IIIIIIIIIIIIIII'IIIIII IIIIIIIIIIIIIIIIIIII illIIIHIIHIIIIIIIIIIHIIIlIIIIIIIII IlllllIlIIlIIH lllllllilll WWII “$11M“ IIIHIIIIIIIIIIIIHI WIIIIIIIIIIIIIIIIIIJIHI'. IIIIIII IIIIIIIIII‘IIIIIIIIIIIII IIllIHIIIIIIlIIJIIlIIHl ulilll
IIII IIIIIIIIIII IIIIIIIIIIIIIIII IIIIIIIIlllllIlIIIIIiIIIIIIIIIIIIlillIIIIIIIIIIIIIIIIIIHIIIIIIIIIIIIIIIIIIlIllIIIHIIIIllII HHIIHIIIII Ililillii IIIllIIIIIIII iIIiIII II'IIINII IIHiililllllllilIIl I'lllllllllllllllllll?"ll IllIIIIiIHiElIlI‘HlHIlIHH-‘l

IIII 8 III IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIII IIIIII HIIIIl WW IIIHII IIIIIIIIIIIIIlIIlIlllllllllillllllﬂllgzaa WHI'IIIHITHHHIHWWHIIIIIII'HIHIIIIIHUIilll'lillll‘llilllllllll'dllll llllllllll lllllllll llllllllﬁllﬂillillllllllllllllﬂillll l
' I IIIIIIIIIIIIII IIIIIl‘I'IIIlIlIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII III' IllIIIlIll Illli I lllHHIlllllIIIiIllHIll Illlllllll'Illl‘ lIIiIlllIlllllIllIIIIIIIIIIll'II’IlIIlII'IIll'IIIiI lIlllI'IlIIIHlI'IIIIlIII

II“ C 1IIIIIIII‘IIII IIIIIlIIiIIllllIIIIIIlliIIIIHIIIIIlIIIllIIIIIIIIIIIIIIIIIIIIIIIIIIUIII'IIIIIIIIIIIIIIH I%%’I§EMHIIIIWIIIMEI'IIIIIIIII .I'lillilil Ella IllIIHIIIH HIIIlIllllllIIIIIII.‘ III‘I III'l III'IIHI' Illlllll IIIIIHHIIIHIIIIIIII. III!” II
IIIIIIIIIIIIIIIII' IIIII rIIIIIIIIIIIIIIIIIllIIIIIIIlIlIIIIIIIIllIIIIIIIIIIIIIIIIIIIIlllll Illllllll llllllllli MHHIIIIHIIIHIIIHIIIII'IIMHIIIR” IllllllillilllllllllHIII‘IIlI‘IIlWII IIIIIIIIIIII ll‘l IHIIII lllIHlI-Illlllllllll

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII'IIIII‘IIIIIIWIIH .. IIHHHI‘IIII IIII'; ‘I‘III IIIIIII I
l I l l I llllill ‘ ' l‘ I

. IIII ‘III =IlIIIIIIIl llIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIlllllllﬂwlll ﬁﬁ‘ﬂlmﬂmﬂmﬂllmlﬂl IlIII'IIIHillI511IIIHIIIllIlIIItIHlIIIllIIIIIIIII‘ III IIIIIII‘II I‘ll llllllllll IIIIIIIIIIIIIH'III

Ialll d IIIIIIIIIIllilllllllIIIIIlIIII‘IIIIIIIIIIIIHIIIII IIlI'IIIIIIIIIIIIHIIIIIIIIIIIIIIIlIIIIIIlIIIIIIIlIIlIIllIiIiIlI Illlllllllllll fllll IIIIIIIHIIIIIIIIIIIIIllliIlIlllllllullillllllllillllll lll' I'll IIIIIIII IIIIIIllIIll‘IIIlIIlllIIIIIlIl

   
  
 

 
 
  
 
 
 

’a‘II VI -lllllllll

I‘III IIIIII'IIIIIII'IIIII‘IIII‘IIIIIIIIIIIIIIIIIIIIIII‘III‘III IIIIIIII IIIIIIIIIIIIIIIIIIIIIIIII 'wllmlwmﬁm mllll PIHNIIHIII IIIIII IIIIIIIIIIIIIIIIIII' ‘IIIII IIII IIIIIIIII IIIIII ‘ . IH‘IIIIIII

II 'III'IIIII IIIII IIIII II'IIII IIIIIII“IIIIIIIIIIIIIIIIIIIII IIaIIII IIII IIII IIIII IIII IIIIIIIII IIIIIIIIIIII IIIIIIIHIIIIIIIIIIII II'II Hill IIIII III lIIIIIIIIIIIIIIIIIIIIWII IIIIImI IlllIIIlIIIIIIIIII‘IIllI‘I III'IIIIIIIIII! I ‘.‘.

‘lla a‘llIl lIII 'IIIIII I|IIIIIIIIIIIIIIIIIIIIiIIIIlIIIIIIIIIIIIIllIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII lilllﬂlll mawmwmmmw IIIIEIIIIIIIIIIIIII IIIIII II‘ ‘ 'III . II

III IIIIIIINHW“”“WIIIIIIIIIIIIIIIIIIIIIIIIIII||IIIIIIIIIIIIIIIIIIIIHIHIIIIIHIIIHIMIHIIMMI HIII Ill %HIHH‘IHIII“H Iil’llllillliill lml“IImum“mumHllimlllllllI IIlI‘IIIIIIIIMIIl IIIIIIIIIII' IIIIIII

I. I 1 III

Ill 0 ‘IIIIII IIIIIIII' IIIIHIIIIIIJ lIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIHHIHHIINHIHIﬁﬁgﬂm HIII%HH%IW%WIIHI%MHW IIHII‘II I III lllllIIIllIIIIIIIIIIIIIIIIIIIIIiIIIIIIIlIIIIIIIII IIIIIIIIIIIII'IIII lIIII‘ IIIa IIIIIIIIIl
IIIII

  

        

IIIIIHI I 'I III IIIII IIIII IIIIIII mi IIIII'III‘IIIIIIIIIIIIIIIIIIIIII!IIIIIIIII'III‘III
'I! IIII1IIII‘.~2 dVlSlOHIIIIIIImlIHIlllII IHHIHIHHIIIIIH' IIIHIIIIIIIIIIIMHIIIH IllIIIIIIII IIIIIIIHI IIIIIIIIIIIIII‘IIIIIII'IIIIIllll'IIIIIIIIIIIIIII
I“? alill"IIIIIIIIIII‘JIIII‘II“aII‘IIIIII'IIIIIlIII'IIIIIIIIIII IIIII IIIIIIIIIIIIIIII ”Hi IIIHI" IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIII IIIIII I
II IIIIIIIIIIIIIIIII IaIaII'IIII‘IIII'IIII‘III‘IIIIIl'IIIIII|II'IlIIIIIIIIIIIIIIIIIIII‘IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHI IIIIII IIIIIIIIIIIIIIIIII ImIIIIIIIIIIIIIIIIIIIIIIIIIiIIIIlIIIIIa'II’IIIIII‘IIIIIIIIIIII'III Ii
IIII‘IIIIIIIIHIIIIa IIIEIIIIIIII‘IIIIIIIIIIIIII'IIII‘IIIIIIHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHIIIIIIIIIIJHIIIHIIIIIIIIIIIHIIIII'IIIIHJ"IIIH‘ZIIIIIIIIIlIIIIIl'I IIIIIIIIIIIIIIIIIIIIIIIIIIIIII :1
ltllllilMllilliliiliIlmlilllillillllllllllllllllllillllllIlliiiullliiliillllIllllillllllliiliulimillm'liliimmllmllﬂlillm “Hill“ Wimmiummwumnumlmwmmmw re. .Uenc

Figure A.12 Experimental Results

185

EXperim en tol Results

Reflected Power Versus Frequency

RI
Ie'
ELF
'l
;e

c:

.t
.e

oi

I .

— 2 .
Fiome 71b n . No Flame

' 0.2 MHZ er division

 

30501]

Frequeni:

Figure A.13 Experimental Results

186

of the air present in the cavity causing a change in the resonant
frequency.

The interaction may seem small but it should remember that the
chemical power of the flame is >1000 watts and only 25 Watts of
microwave power was available. In addition it should be remembered that
the region of the plasma in the flame is very small and may not be

oriented in the most ideal manner for microwave coupling.

A.4 CONCLUSIONS

The result of the experiments were inconclusive. However, they
demonstrated important methods and experimental setups that with further
refinement could demonstrate microwave interaction with a flame.
Earlier results presented in the literature review support that this is
possible and should be studied. Additional experiments need to
conducted that study the chemistry and energy changes that occur with
the addition of a microwave field. Studies of the basic microwave-flame
interaction would also be beneficial to the understanding to combustion

flames and microwave discharges.

L IST OF REFERENCES

LIST OF REFERENCES

Hhitehair, S., Asmussen, J., and Nakanishi, S., "Demonstration of a
New Electrothermal Thruster Concept," Applied Physics Letters, vol.
44, p. 1014-1016, 1984.

Hhitehair, S., Asmussen, J., and Nakanishi, S., "Microwave
Electrothermal Thruster Performance in Helium Gas," accepted for
publication in Journal of Propulsion and Power.

Hhitehair, S. and Asmussen, J., "The Generation and Maintenance of
Stable High Pressure Microwave Arcs," presented at the Fifth Topical
Conference on Radio Plasma Heating ,February 21-23, 1983, Madison,
HI, also published in the conference proceedings, Proceedings of the
Fifth Topical Conference in Radio Frequency Plasma Heating, p. 296-
299.

Asmussen, J. and Hhitehair, S., "Measurement of the Electromagnetic
Field Pattern Surrounding Atmospheric Microwave Discharges,"
presented at the 1983 IEEE International Conference on Plasma
Science, San Diego, CA, May 23-25, 1983.

Asmussen, J. and Whitehair, S., "Performance Measurements on a
Microwave Electrothermal Engine," presented at the 1984 IEEE
International Conference on Plasma Science, St. Louis, MD, May 14-
16, 1984.

Asmussen, J., Hhitehair, S., and Nakanishi, S., "Experiments with a
Microwave Electrothermal Thruster Concept," AIAA/JSASS/DGLR 17th
International Electric Propulsion Conference, IEPC 84-74, May 1984,
Tokyo , Japan .

Hhitehair, S., Asmussen, J. and Nakanishi, 8., "Recent Experiments
with Microwave Electrothermal Thrusters," presented at the
AIAA/DGLR/JSASS 18th International Electrical Propulsion Conference,
Alexandria, VA, AIAA-85-2051, Sept. 30-Oct. 2, 1985.

Hhitehair, S., Frasch, L. and Asmussen, J., "Experimental
Investigation of Microwave Electrothermal Thruster Configurations,"
presented at the 13th IEEE International Conference on Plasma
Science, Saskatoon, Saskatchewan, Canada.

Hhitehair, S. and Asmussen, J., "An Electrodeless Microwave
Plasmatron," disclosure to Michigan State University, April 18,
1983.

187

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

188

Hhitehair, S., Asmussen, J., Nakanishi, S. and Malinkey J.,
"Microwave Electrothermal Thruster," disclosure to NASA Lewis
Research Center, Oct. 25, 1983.

Fredericks, R., " An Experimental and Theoretical Study of a
Resonantly Sustained Plasma in Microwave Cavities," Ph.D.
Dissertation, Michigan State University, 1971.

Mallavarpu, R., "An Investigation of the Electromagnetic Behavior of
a Microwave Plasma Source Over a Hide Range of Pressures and Flow
Rates," Ph.D. Dissertation, Michigan State University, 1976.

Rogers, J.,"Properties of Steady-State, High Pressure Argon
Discharges," Ph.D. Dissertation, Michigan State University, 1982.

Sutton, G.P., Rocket Propulsion Elements, John Wiley 8 Sons, New
York, 1976.

Garrison, P.H., "Advanced Propulsion for Future Planetary
Spacecraft," Journal of Spacecraft and Rockets, vol. 19, p. 534-538,
1982.

Hawkins, C.E. and Nakanishi, S.," Free Radical Propulsion Concept,"
NASA TM 81770, 1981.

Chapman, R., Filpus, J., Morin T., Snellenberger, R., Asmussen J.,
Hawley M. and Kerber R., "Microwave Plasma Generation of Hydrogen
Atoms for Rocket Propulsion," Journal of Spacecraft and Rockets,
vol. 19, p. 579, Nov.-Dec. 1982.

Jahn, R.G., Physics ofgglectric Propulsion, McGraw-Hill, New York,
1968.

Page, R.J., Halbach, C.R. and Short, R.A., "3-kH Concentric Tubular
Resistojet Performance," Journal of Spacecraft and Rockets, vol. 11,
p. 1669-1674. 1966.

Page, R.R. and Stoner, H.A., "The Design and Technology Development
for a 150 ML8 Resistojet for H2 or NH3," AIAA/JSASS/DGLR 16th
International Electric Propulsion Conference AIAA-82-1949, November
17-19, 1982.

Zafran, S., and Jackson, 8.,"Electrothermal Thruster Diagnostics,"
Vol. I and II., TRH Space Technology Group, NASA CR 168174, May,
1983.

Jackson, F.A., et al, "An Operational Electrothermal Propulsion
System for Spacecraft Reaction Control," AIAA Paper 66-213, 1966.

Schoeck, P.A., "Physical Phenomena and Technical Problems of Arcjet
Engines," Journal of the British Interplanetary Society, vol. 19, p.
337-349, 1963-64.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

189

John, R.R., Bennett, 5. and Connors, J.F., "Arcjet Engine
Performance: Experiment and Theory," AIAA Journal, vol. 1, p. 2517,
1963.

McCaughey, D.J., "Development of a Plasmajet Rocket Engine for
Attitude and Orbit Control," Report No. FR-112-651, Plasmadyne
Corp., June 1964.

Nallner, L.E. and Czika, J., "Arc-Jet Thruster for Space
Propulsion," NASA TN D-2868, June 1965.

Ducati, A.C., Giannini, G.M. and Muehlberger, E, "Experimental
Results in High Specific Impulse Thermo-Ionic Acceleration," AIAA
Journal, vol. 2, p. 1454, 1964.

Kantrowitz, A., "Propulsion to Orbit by Ground-Based Lasers,"
Astronautics and Aeronautics, vol. 10, p. 74-76, 1972.

Glumb, R.J. and Krier, H., "Concepts and Status of Laser-Supported
Rocket Propulsion," Journal of Spacecraft and Rockets, vol. 21, p.
70-79, 1984.

Franzer, D.L., "Continuous Laser-Sustained Plasmas," Journal of
Applied Physics, vol. 49, p. 1727-1731, 1973.

Gulati, A. and Merkle, D.L., "Absorption of Electromagnetic
Radiation in an Advanced Propulsion System," Journal of Spacecraft
and Rockets, vol. 21, p. 101-107, 1984.

Jackson, J.P. and Nielson, P.E., "Role of Radiative Transport in the
Propagation of LSC Haves," AIAA Journal, vol. 12, p. 1498-1501,
1974.

Reed, T.8., "Induction-Coupled Plasma Torch," Journal of Applied
Physics, vol. 32, p. 821-824, 1961.

Eckert, H.V., and Pridmore-Brown, D.C., "Temperature Profiles in
Argon Induction Plasmas: Theory and Experiment," Journal of Applied
Physics, vol. 42, p. 5051, 1971.

Reed, T.8., "Growth of Refractory Crystals Using the Induction
Plasma Torch," Journal of Applied Physics, vol. 32, 1961.

Jones, R.M., "Comparison of Potential Electric Propulsion Systems
for Orbit Transfer," Journal of Spacecraft and Rockets, vol. 16, p.
88-95, 1984.

Petrash, D.A., "Station Keeping Hill Eat Energy On a New Scale,"
Astronautics and Aeronautics, vol. 21, p. 64-65, 1983.

Kaufman, H.R. and Robinson, R.S., "Electric Thruster Performance for
Orbit Raising and Maneuvering," Journal of Spacecraft and Rockets,
vol. 21, p. 180-186, 1984.

39.

40.

41.

42.

43.

44.

45.

46.

47a

‘8.

49.

50.

51.

52.

53.

190

Berkopec, F.D. and Stone, J.R., "NASA Electric Propulsion
Technology," AIAA paper 85-1999, 1985.

Gilmour, A.S., Microwave Tubes, Artech House, 1986.

McMillan, R.M., "Near-Millimeter-Have Sources of Radiation,"
Proceedings of the IEEE, vol. 73, p. 86-107, 1985.

Crapuchettes, P.H., "Magnetrons as Generators of Microwave Power,"
Microwave Power Engineering, vol. 1, p. 18-42, Academic Press, New
York, 1968.

Brown, H.C., "The Amplitron," Microwave Power Engineering, vol. 1,
p. 84, Academic Press, New York, 1968.

Neugebauer, H., and Mihran, T., "A Ten State Electrostatic Depressed
Collector for Improving Klystron Efficiency," IEEE Transactions on
Electron Devices, vol. 19, 1972.

M/A-COM Microwave Power Devices, Inc., "Update: RF/Microwave Power".

Nalos, E. J., "New Developments in Electromagnetic Energy Beaming,"
Proceedings of the IEEE, vol. 66, p. 276-289, 1978.

Frasch, L. and Asmussen, J., "Electrothermal Propulsion of
Spacecraft with Time-Varing Electromagnetic Energy," AIAA 18th
International Electric Propulsion Conference, Alexandria VA, Sept.
30 to Oct. 2, 1985.

Asmussen, J. and Rogers, J., "An Assessment of the Potential
Application of Microwave Discharges to Electric Propulsion Engines,"
MSU/NASA Lewis Research Center Conference on Advanced Propulsion
Concepts Using Time Varying EM Fields, Michigan State University,
Feb. 23-25, 1982.

Babat, G.I., Electrodeless Discharges and Some Allied Problems,"
Journal of the Institute of Electrical Engineers, London, vol. 94,
p. 27-37, 1947.

Asmussen, J., Mallavarpu, R., Hamann, J.R., and Park, H.C., "The
Design of a Microwave Plasma Cavity," Proceedings of the IEEE, vol.
62(1), p. 109-117, January, 1974.

Fredericks, R.M. and Asmussen, J., "High-density Resonantly
Sustained Plasma in a Variable-Length Cylindrical Cavity," Applied
Physics Letters, vol. 19, p. 508-510, 1971.

Halverson, S.L. and Hatch, A.J., "Frequency-Shifting Method of
Exciting a Plasma in a Resonant Cavity," Applied Physics Letters,
vol. 14, p. 79, 1969.

Moisan, M., Beaudry, C. and Leprince, P., "A Small Microwave Plasma
Source for Long Column Production Without Magnetic File," IEEE
Transactions on Plasma Science, vol. 3, p. 55, 1975.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

191

Fritz, R. and Asmussen, J., "Experimental Characteristics of RF
Surface Have Generated Plasma Sources," IEEE International
Conference on Plasma Science, Montreal, Canada, June 1979.

Cobine, J.D. and Hilbur, D.A., "The Electric Torch and Related High
Frequency Phenomena," Journal of Applied Physics, vol. 22, p. 835-
841. 1951.

Moriarty, J.J. and Brown, V.M., "Toroidal Microwave Discharge
Heating of Gas," Journal of Microwave Power, vol. 3, p. 180-186,
1968.

Rogers, J. and Asmussen, J., "The Experimental Production of
Microwave Plasma Filaments and Discharges at High Pressure,"
proceedings 1981 IEEE International Conference on Plasma Science,
May 18-20, 1981.

Miyake, S., Takeuchi, S. and Arata, T., "The Experimental
Investigation of Coaxial Microwave Plasmatron in Nitrogen Gas,"
Japanese Journal of Applied Physics, vol. 13, p. 296-305, 1974.

Mallavarpu, R., Asmussen, J., and Hawley, M.C., "Behavior of a
Microwave Cavity Discharge over a Hide Range of Pressures and Flow
Rates," IEEE Transactions on Plasma Science, vol. PS-6(4), p. 341-
354, December, 1978.

Asmussen, J. and Root, J., "Characteristics of a Microwave Plasma
Disk Ion Source," Applied Physics Letters, Vol. 44, p. 396-398,
1984.

Asmussen, J. and Root, J., "Ion Generating Apparatus and Method for
the Use There Of," U.S. Patent. No. 4 507 588, Mar. 26, 1985.

Root, J. and Asmussen, J., "Experimental Performance of a Microwave
Cavity Plasma Disk Ion Source," Review of Scientific Instruments,
vol. 56, p. 1511-1519, 1985.

Ramo, S., Hhinnery, J.R., and Van Duzer, T., Fields and Waves in
Communicationsgglectronics, lst ed., J. Wiley and Sons, New York,

.1972.

Fredericks, R.M., and Asmussen, J.,"Retuning and Hysteresis Effects
of a rf Plasma in a Variable Size Microwave Cavity," Journal of
Applied Physics, vol. 42, p. 3647-3649, 1971.

MacDonald, A.D. and Tetenbaum, S.J., "High Frequency and Microwave
Discharges," Gaseous Electronics, Hirsch, M.N. and Oskam, H.J.
editors, vol. 1, Academic Press, New York, 1978.

Marec, J., Bloyet, E., Chaker, M., Leprince, P. and Nghiem P.,
"Microwave Discharges," Electrical Breakdown and Dischacggs in
Gases, Kunhart, E.E. and Luessen H.L. editors, Part B, Plenum
Publishing Corp., 1983.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

192

Pfender, E., "Electric Arcs and Arc Gas Heaters," Gaseous
ggectronics, Hirsch, M.N. and Oskam, H.J. editors, vol. 1, Academic
Press, New York, 1978.

Chapman, R., Morin, T., Finzel, M. and Hawley, M.C., "Measurement of
Energy Distribution in Flowing Hydrogen Microwave Plasmas," Journal
of Spacecraft and Rockets, vol. 22, p. 626, Nov.-Dec. 1985.

Cherrington, 8.8., Gaseous Electronics and Gas Lasers, Pergamon
Press, Oxford, 1979.

Elenbaas, H., ight Sources, Philips Technical Library, Crane,
Russak & Company, New York, 1972.

Kapitza, P.L., "Free Plasma Filament in a High Frequency Field at
High Pressure," Soviet Physics JETP, vol. 30, p. 973-1224, 1970.

Kapitza, P.L. and Filimonov, 5.1., "Apparatus for the Production of
a Free Plasma Filament. Determination of the Current and Resistance
of the Filament," Soviet Physics. JETP, vol. 34, p. 542-553, 1972.

Dymshits, B.M. and Koretskii, Y.P., "Temperature of a Free Plasma
Filament in a High Frequency Field at High Pressure," Opt. Spect.,
vol. 33, p. 17-18, 1972.

Arata, Y.,Miyake, S. and Kobayashi, A., "Research of a Stationary
High Power High Power Microwave Plasma at Atmosphere Pressure,"
Journal of the Physical Society of Japan, vol. 40, p. 1456-1461,
1976.

Arata, Y., Miyake, S. and Kobayashi, A., "High Power Microwave
Discharge in Atmospheric Hydrogen Gas Flow," Journal of the Physical
Society of Japan, vol.44, p. 998-1003, 1978.

Rogers, J. and Asmussen, J., "Properties of Steady State, High
Pressure Microwave Generated Arc Discharges in Argon Gas," 1982 IEEE
International Conference on Plasma Science, Ottawa, Canada, May 17-
19, 1982.

Hubert, J., Moisan, M. and Ricard, A., "A New Microwave Plasma at
Atmospheric Pressure," Spectrochimica Acta, vol. 338, p. 1-10, 1979.

Batenin, V.M., Zrodnikov, V.S., Roddatis V.K. and Chinnov, V.F.,
"Spectroscopic Investigation of a Microwave Discharge in Hydrogen,"
High Temperature, vol. 13, p. 250-257, 1975.

Batenin, V.M., Zrodnikov, V.S. and Chinnov, V.F., "Microwave
Discharge in Helium," High Temperature, vol. 14, p. 186-188, 1976.

Bandel, H.H. and MacDonald, A.D., "Microwave Breakdown in Air Plus
H20," Journal of Applied Physics, vol. 41, p. 2903-2904.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

193

Mertz, S.F., Asmussen, J. and Hawley M.C., "An Experimental Study of
Reactions of CO and H2 in a Continuous Flow Microwave Discharge
Reactor," IEEE Transactions on Plasma Science, vol. PS-2, p. 297-
307. 1974.

Hard, M.A.V.,"Potential Uses of Microwave to Increase Internal
Combustion Energy Efficiency and Reduce Exhaust Pollutants", Journal
of Microwave Power, vol. 12(3), p. 187-198, 1977.

Hard, M.A.V., and Hu, T.T.,"A Theoretical Study of the Microwave
Heating of a Cylindrical Shell, Flame-Front Electron Plasma in an
Internal Combustion Engine", Combustion and Flame, vol. 32, p. 57-
71, 1978.

Frimstrom, R.M. and Hestenburg, A.A., Elgme Structure, McGraw-Hill
Book Company, 1965.

Gaydon, A.G. and Holfhard, H.G., Flames, Their Structure, Radiation
and Temperature, Barn and Noble Inc.,1970.

Glassman, I., Combustion, Academic Press, 1970.

Jaggers, H.C. and Von Engel, A.," The Effect of Electric Fields on
the Burning Velocity of Various Flames", Combustion and Flame, vol.
16, p. 275-285, 1971.

Tewari, G.P., and Hilson J.R.," An Experimental Study of the Effects
of High Frequency Electric Field on Laser Induced Flame
Propagation", Combustion and Flame, vol. 24, p. 159-167, 1975.

MacLatchy, C.S., Clements, R.M., and Smy, P.R.,"An Experimental
Investigation of the Effect of Microwave Radiation on a Propane-Air
Flame", Combustion and Flames, vol. 45, p. 161-169, 1982.

Clements, R.M., Smith, R.D., and Smy, P.R., "Enhancement of Flame
Speed by Intense Microwave Radiation," Combustion Science and
Technology, vol. 26, p. 77-81, 1981.

Hard, M.A.V.,"Interaction of Propane-Air Laminar Flame Plasmas with
the Microwave TM010 Mode Excited in a Cylindrical Combustion Bomb
Cavity", Journal of Microwave Power, vol. 14(3), p. 241-259, 1979.

Hard, M.A.V.,"Microwave Stimulated Combustion", Journal of Microwave
Power, vol. 15(3), p. 193-202, 1980.

 

"1111111111 1111“