Low Power Ground-Based Laser Illumination for Electric Propulsion Applications

¢¢ " O<,... #

_:'.i>37.5 ,, /" NASA Contractor Report 194444 IEPC 93-208 28P

Michael R. LaPointe and Steven R. Oleson Sverdrup Technology, Inc. Lewis Research Center Group Brook Park, Ohio

(NASA-CR-194444} LOW POWER N9_-23555 GROUND-BASEO LASER ILLUMINATION FOR ELECTRIC PROPULSION APPLICATIONS Final Report (Sverdrup Techno|ogy) Unclas January 1994 28 p

G3/20 0203577

Prepared for Lewis Research Center Under Contract NAS3-25266

National Aeronautics and Space Administration

LOW POWER GROUND-BASED LASER H,I,UMINATION FOR ELECTRIC PROPULSION APPLICATIONS

Michael R. LaPointe and Steven R. Oleson Sverdrup Technology, Inc. LeRC Group, Brook Park, OH 44142

ABSTRACT l.,p specific impulse (s) LACE Low-Power Atmospheric Compensation

A preliminary evaluation of low power, ground-based Experiment LEO Low Earth Orbit laser powered electric propulsion systelns is presented. A review of available and near-term laser, LLNL Lawrence Livermore National Laboratory photovoltaic, and adaptive optic systems indicates that Mo total initial spacecraft mass (kg) approximately 5-kW of ground-ba_d laser power can M v expended propellant mass {kg) be delivered at an equivalent l-sun intensity to an MIRACL Mid-lnfraRed Advanced Chemical Laser orbit of approximately 2000 kin. Laser illumination MPD magnetoplasmadynamic thruster at the proper wavelength can double photovoltaic Ny extended satellite life in orbit (y) array conversion efficiencies compared to efficiencies Nd:YAG neodymium:yttrium aluminum garnet obtained with solar illumination at the same intensity, PL laser power (W) allowing a reduction in alray mass. The reduced PIT pulsed inductive thruster array mass allows extra propellant to be carried with PPT pulsed plasma thruster initial beana radius (m) no penalty in total spacecraft mass. The extra propel- ro lant mass can extend the satellite life in orbit, allow- r._v,,, final beam radius (m) silicon ing additional revenue to be generated. A trade study Si using realistic cost estimates and conservative ground SPT stationary, plasma thruster station viewing capability was performed to estimate SWAT Short Wavelength Adaptive Techniques the number of communication satellites which must Z propagation distance (m) be illuminated to make a proliferated system of laser or= molecular absorption coefficient (nf _) aerosol absorption coefficient (m") ground stations economically attractive. The required cq number of satellites is typically below that of pro- molecular scattering coefficient (m q) aerosol scattering coefficient ( m"_) posed cornmunication satellite constellations, indicating that low power ground-based laser beaming may atmospheric attenuation coefficient (m _) be commerciaUy viable. However, near-term advanc- AV velocity change, delta-V (m/s) es in low specific mass solar arrays and high energy Z wavelength (m) density batteries for LEO applications would tender T atmospheric transmittance the ground-based laser system impracticable. l -sun equivalent l-sun intensity (I.38 kW/nt)

NOMENCLATURE I. LNTRODUCTION ACE Atmospheric Compensation Experiment APSA Advanced Photovoltaic Solar An'ay Laser power beaming has been advocated for a AVLIS Atomic Vapor Laser Isotope Separation variety of spacecraft power, propulsion, and communication applications): Early research focnsed on the C_ cost of laser groutv, l station ($) DF deuterium fluoride design of multimegawatt space-based lasers, and the FL:L free electron laser development of suitable high power conversion tech- g acceleration due to gravity (9.8 nVs 2) nologies? Orbiting laser power stations were pro- GaAs gallium arsenide posed as a methtxl to deliver power to a multitude of GEO Geosynchronous Earth Orbit (3.6xl07 m) reusable orbit transfer vehicles, whose reuse could HF hydrogen fluoride amortize the expense of a space transportation infra- I intensity (W/m _) structure. s High power laser systems were envisioned 1o initial beam intensity (W/m 2) to occupy stationary lunar orbits, providing continuous power for bases, rovers, and ltmar mining operations, 6 radius after propagating a distance Z, and ,_ is the and were proposed as a method to deliver power from laser wavelength. The final beam or spot radius in Mars orbit to a base on the Martian surface as part of Equation i corresponds to the first zero in the diffrac- an ambitious program of planetary exploration. 7 tion pattern at the receiver, which contains 84% of the initial beam energy. For a laser wavelength of 1.06 Although attractive in terms of potential benefits, lum and an initial beam radius of 0.5 m, the spot several hurdles must be overcome in the design, radius at a distance of 500 km is approximately 0.65 development, and operation of muitimegawatt space- m due to diffraction, which corresponds to a beam based laser stations. High power chemical lasers are expansion half-angle (r_/Z) of approximately 1.3 well developed and routinely used for industrial and vrad. In the absence of atmospheric effects, larger military applications, t'_ but the need to continually initial beam radii or smaller laser wavelengths could replenish the chemical reactants would significantly be used to further reduce the beam radius at the increase the operating cost of a laser station in orbit. receiver. Solar-pumped lasers, which use solar radiation to induce lasant population inversions, have been suc- The effect of atmospheric turbulence on beam propa- cessfully demonstrated at sub-kW power levels, t'9 gation is a function of the atmospheric coherence However, a tremendous amount of work remains distance, _ which is a measure of the lateral distance before these concepts can provide the high power, over which atmospheric fluctuations do not signifi- closed-cycle, autonomous operation required for the cantly effect beam propagation. The atmospheric multimegawatt mission applications proposed above. coherence distance depends upon the path length, the Nuclear-pumped lasers, which utilize fission frag- beam propagation angle, and the refractive index ments to excite a iasant gas in a reactor core, have structure "constant', a complex and decidedly non- also been developed and tested at low power levels?'_° constant variable which approximates the strength of However, continual postponements in the development the atmospheric turbulence. For wavelengths of and deployment of space nuclear power systems make interest to laser power beaming, the coherence dis- it unlikely that a space-based nuclear-pumped laser tance is on the order of 0. I m, '4 which roughly corre- will be available in the near future. Free electron sponds to the maximum initial beam diameter that can lasers, in which coherent radiation is extracted from be propagated through the atmosphere without serious the periodic oscillations of high energy electron degradation. For initial beam diameters smaller than beams, I1 can generate high power levels but require the atmospheric coherence length, the final spot size significant development before they can provide is governed primarily by diffraction. For larger initial reliable, long-term operation in the remote space beam diameters, the final spot size is governed environment. primarily by atmospheric turbulence. An initial beam diameter larger than the atmospl'_eric coherence Many of the issues associated with using space-based distance will not decrease the beam expansion caused lasers are removed by keeping the laser stations on by atmospheric turbulence. Instead, the beam will the ground, where operating power is readily available spread as if the effective initial beam diameter were and the lasers are accessible for maintenance and equal to the atmospheric coherence distance. upgrade. Enthusiasm for ground-based power beaming was initially tempered by the difficulties associat- The final spot radius of a beam propagating through ed with propagating high power laser beams through the atmosphere is given by the addition of the beam the atmosphere. Atmospheric turbulence and temper- expansion due to diffraction and the beam expansion ature fluctuations change the atmospheric index of due to atmospheric turbulence. For the parameters refraction along the beam propagation path, _2produc- used in the above example, an initial beam radius ing distortions which can spread the beam and dra- equal to half the atmospheric coherence distance matically alter the intensity profile at the receiver. In yields a spot radius of approximately 6.4 m. corre- the absence of atmospheric distortion, the beam will sponding to an expansion half-angle of about 13 _trad spread as it propagates due to diffraction at the beam due solely to atmospheric turbulence. The final spot source such that: xadius for the example is then given by:

r,m _ (0.61)(5x10 -_m)(I.O6xl0 -_ m)/(0.5 m) r,r,, = 0.61 Z )_ (m) (1) (2) + (5x10 "_m)(l.3xl0 "-_rad) - 7 m r0 where ro is the initial beam radius, r,_,, is the beam which is significantly larger than the 0.65 m spot radius calculated for diffraction effects alone. A cal signals and used to drive the deformable mirror. which compensates for the phase distortions and larger initial beam radius would reduce tile diff,'active flattens the incoming wave. With the use of adaptive component of beam spreading, but it would not alter optics, the inaage resolution can be nearly diffraction- the significantly larger beam expansion caused by limited. In addition to natural stars, laser guide stars atmospheric turbulence. The larger receiver area have been used to provide point-like sources for required to collect the expanded beam after it travers- es the atmosphere increases the mass of the space- adaptive optics when natural stars are too weak or too far from the desired point of observation. Techniques craft, and mitigates file potential advantages of laser using Rayleigh backscatter _7and laser illumination of power beaming. the Earth's sodium layer _8 have provided effective

In addition, wavefi'ont phase changes imparted by synthetic beacons for the adaptive optics correction of astronomical objects. atmospheric fluctuations can produce localized regions of high beam intensity at the receiver, often of sufficient magnitude to damage or degrade receiver Similar adaptive optic techniques may be used to performance. Other concerns associated with high propagate laser beams through the atnlosphere. A natural star or synthetic beacon is nsed to provide a power beam propagation include thermal blooming, t'_ in which a fraction of the transmitted laser energy is reference signal for the wavefront sensor. The wavefi'ont sensor controls an active mirror, which absorbed by the atmosphere along the propagation deforms to fit the phase profile of the incoming path. The atmosphere becomes slightly heated, distorted wavefront. The laser beam to be propagated changing the refractive index and creating a negative lens along the beam path length. The bearn spreads through the atmosphere is reflected from the de- formed mirror, which imparts a phase distortion to the radially due to the negative atmospheric lensing, and may be further distorted by the asymmetric heat flow outgoing wavefront that is the approximate conjugate associated with atmospheric winds or by the laser of the phase distortions accumulated by the reference beam slewing tl'u'ough the atmosphere. Without a signal on its downward path tlu'ough the atmosphere. suitable technique to compensate for these atmo- As the predistorted outgoing wavefront propagates spheric aberrations, ground-based laser power beam- back along the same (or nearly the same) path ing is not a particularly attractive option compared to through the atmosphere, the phase distortions are conventional methods of spacecraft power and propul- reversed, and a nearly diffraction-limited planar sion. Fortunately, methods to correct such atmo- wavefront emerges from the atmosphere. For moving targets, the reference signal must be placed ahead of spheric distortions exist in the form of adaptive the target so that the reference signal and return beam optics. propagate along approximately the same path through Although still developmental, adaptive optics have the atmosphere. For satellites in non-geosynchronous been successfully used to correct the atmospheric orbit, this requires the use of beacon lasers or retrote- distortion of astronomical objects and to beam laser flectors placed on extended booms in the direction of energy through the atmosphere? _ Adaptive optic motion (Figure 2), or synthetic Rayleigh or sodium systems for astronomical applications typically consist layer beacons created at the correct point-ahead of a telescope receiver, a wavefront sensor, an active distance in the atmosphere so that the return laser or deformable mirror, and a control system to convert beam intercepts the main satellite body (Figure 3). the output from the wavefront sensor into signals which control the deformable mirror (Figure 1). The ability of adaptive optics to compensate for Light from a target star is collected by the telescope, atmospherically induced laser beam aberrations has and reflected from a fast steering tilt mirror to a been demonstrated in a number of successful ground- deformable mirror. A portion of the light is sent from to-space beam propagation experiments, performed the deformable mirror to an imaging camera, and the over the past decade by the Massachusetts Institute of remainder is sent to a wavefront sensor which mea- Technology Lincoln Laboratories under the aegis of sures the atmospherically induced phase perturbations the Department of Defense? 9 Part of the Atmo- across the telescope aperture. Because the star is an spheric Compensation Experiment (ACE) test series used a retroreflector carried aboard the space shuttle effective point source, the incoming light ought to be a plane wave. The phase perturbations measured by Discovery and a ground-based, 60-cm diameter the wavefront sensor correspond to variations from deformable mirror to perform preliminary adaptive the expected plane wave distribution. The phase optic compensation experiments. An uncon-ected laser beam was used to illuminate the retroreflector, perturbation measurements are converted into electri andthe reflected signal was successfitlly corrected by compensation experiments, providing tile first demon- the ACE adaptive optics. The experiment demonstrat- stration of the synthetic beacon technique for compen- ed atmospheric compensation of a dynamic target, but sated ground-to-satellite beam propagation. 20 The did not actually compensate an outgoing laser beam. I,ACE satellite was decommissioned in February. A series of subsequent ACE tests were performed 1993. wilh sounding rockets, launched to altitudes of approximately 600 kmfl Each rocket carried a With a viable method to propagate ground-based laser retroreflector to provide a synthetic beacon for the energy through the atmosphere, high power lasers adaptive optics, and a linear array of detectors to have again been advocated for space power and measure beam compensation. The retroreflectors propulsion applications. Potential missions include the were illuminated with laser light at 488 nm, providing illumination of geosynchronous satellite solar arrays a synthetic beacon for the wavefront sensor. The during eclipse periods, 2_the illumination of thermal or wavefront information was used to control a deform- electric propulsion orbital transfer vehicles, z2J-_and the able mirror, and a second laser at 514 nm was reflect- illumination from Earth of a lunar base during the 14- ed from this mirror and detected by the passing day lunar night. _4 Each of these applications require rocket. The detector arrays recorded a dramatic incr- sustained laser powers of several hundred kilowatts to ease in beam irradiance when the outgoing laser beam tens of megawatts, and as noted above the necessary was compensated for atmospheric aberrations by the laser systems are still being developed. adaptive optics, and the tests were the first to successfully demonstrate the atmospheric compensation of a This paper presents a first order consideration of laser beam propagated from the ground-to-space. potential mission opportunities which can take advantage of adaptive optics acting in concert with avail- Other programs at the MIT Lincoln Laboratory able laser and power conversion systems. The included the Short-Wavelength Adaptive Techniques following section presents an overview of available (SWAT) test series, 2° which used atmospheric Ray- and near-term technologies for laser beam formation, leigh backscatter to create a synthetic beacon for propagation, conversion, and utilization for spacecraft compensated astronomical observations. Subsequent propulsion. Section I11 outlines potential applications. SWAT tests demonstrated retroreflector and synthetic including a look at low laser power ground-based beacon compensation techniques for power beaming beaming for conunercial satellite constellations. The to a satellite in LEO. In February 1990, the Low- paper concludes with a brief summary of results, and Power Atmospheric Compensation Experiment suggestions tbr further research. (LACE) research satellite was placed into a 547 km, 43° inclination circular orbit. A corner cube array, located on an extendable boom and illuminated with H. TECHNOLOGY REVIEW an uncompensated ground-based laser, was used to provide a return signal for sensing the atmospheric The general requirements for ground-based laser distortion. The reflected signal was used by the power beaming include the ability to transmit the ground-based adaptive optics to correct a second laser energy through the atmosphere without signifi- outgoing laser beam. This corrected beam was cant losses, the efficient conversion of the laser power detected by an array of silicon photodectors distrib- to electrical power via photovoltaic conversion or the uted in a 2-dimensional pattern on the satellite body. absorption of the laser power for thermal plopulsion, Real-time detector array data were used to evaluate and at least a competitive ability to perform a given the effectiveness of the compensation, and to drive a mission. The following sections present an overview pointing loop which kept the outgoing beam centered of atmospheric propagation issues, commercial and on the array. The duration of an overhead pass, from near-term laser md photovoltaic array technologies. a sighting elevation of 45 ° ascending to 45" descend- and candidate electric propulsion systems which might ing, was 120 to 150 s, corresponding to a laser slew be used for low power mission applications. rate of approximately 10 mrad/s. The experiments, performed over a 15 month period, demonstrated the Atmospheric Propagation Issues. ability of the adaptive optics to compensate a ground- based laser beam for low earth orbit satellite applica- In addition to the diffractive beam spreading dis- tions. As part of the SWAT program, a synthetic cussed above, both linear and nonlinear atmospheric beacon created in the atmosphere with Rayleigh effects hamper the efficient propagation of laser backscatter was successfully used for LACE satellite energy front ground to space. Some lhlear effects,

4 suchasatmospheric turbulence and fluctuations in the cial laser types exist which are of interest for power atmospheric refractive index, can be corrected using beaming applications. The following brief descrip- adaptive optics. Other linear effects, such as beam tions are compiled from the 1986 Laser Guidebook, 2_ scattering and absorption by molecules and aerosols, the 1990 Lasers and Optronics Buyer's Gitide, 27 and or beam attenuation due to inclement weather, remove the 1993 Laser Focus World Buyer's Guide, 2s with directed energy from the beam and cannot be correct- additional information provided by the cited referenc- ed using adaptive optics. Nonlinear effects, such as es. The descriptions are not intended to provide a co- thermal blooming and air breakdown, require laser mprehensive tutorial on all possible laser systems of beam intensities on the order of 10t-10 s W/cm 2. Such interest, but rather serve to illustrate the general laser intensities are not expected to occur for the low nature of commercially available lasers which might power cw lasers considered in this study, and can be be considered for near-term power beaming applica- avoided in pulsed laser systems by keeping the pulsed tions. beam intensities below these approximate thresholds. Dye lasers use a fluorescent organic dye in a liquid Beam scattering and absorption by molecules and solvent as the lasant medium. Intense illumination by aerosols present the most serious challenge to beam a separate source I flashlamp, ion laser, copper vapor propagation through the atmosphere. The atmospheric or Nd:YAG laser) excites the dye molecules to transmittance (_) is defined as: _s produce a population inversion. The dye then under- z goes stimulated emission to produce a laser beam. r = I(z) = exp[-J'_ dz] _3) Dye lasers are tunable from roughly 300 nm to 1000 I0 e nm, depending upon the dye. Depending upon the where l(z) is the beam intensity after propagating a pump source, average or cw powers can vary from distance z through the aunosphere, Io is the initial several tens of watts for commercial units to kilowatt- beam intensity, and i_ is the atmospheric attenuation class special order lasers. Beam diameters may range coefficient, given by: from slightly less than 1 mm up to 20 ram, with beam divergence angles of 0.3 mrad to 6 mrad. Laser dye = or,, + oq + I_®+ 13, ira") (4) solutions have limited lifetimes ranging from several hours to several months, and flashlamp lifetimes are where oq, is the molecular absorption coefficient, or, limited from 104-106 shots before replacement is is the aerosol absorption coefficient, 15, is the molecu- required. The laser systems are not particularly lar scattering coefficient (due to Rayleigh scattering), robust, and rough handling could damage the liquid and 13,is the aerosol scattering coefficient (due to Mie flow system and misalign the optics. Laser dyes and scattering). Molecular absorption is highly dependent solvents can be toxic, and most are flammable both in upon wavelength, and its proper evaluation requires a liquid and vapor form. Typical costs for the higher detailed knowledge of the spectroscopic parameters of power commercial dye lasers range from $50,000 to thousands of absorption lines in the atmosphere. $100.000. Molecular scattering is in general only important for ultraviolet radiation. Aerosol scattering is generally Noble gas ion lasers use argon, krypton, or a combi- more important than aerosol absorption. Aerosol nation of the two gases. A high current discharge is attenuation is a slowly varying function of wave- used to ionize the gas and induce the required popula- length, and is generally less important at longer wave- tion inversions. The lasers operate in continuous- lengths. Figure 4 shows a low resolution plot of wave mode. but may be modelocked for pulsed atmopsheric transmittance versus wavelength from 0- operation. Multiline argon-ion laser output powers 15 microns. Numerous atmospheric absorption lines may reach several tens of watts, with single-line exist which are not shown on the low resolution operation at lower power levels. The less efficient plot, z_ and atmospheric transmission models are krypton-ion lasers can produce up to a few watts of continually being upgraded to better resolve the multiline power. Argon-ion lasers emit at several effects of narrow band absorption on atmospheric wavelengths from 351-528 rim, with main emission laser beam propagation. lines at 488 nm and 514.5 nm. Krypton-ion lasers emit several lines between 350-800 nm, with a main Candidate Laser Systents. line at 647.1 rim. The lasers are long-lived, with lifetimes exceeding several thousands of hours. A variety of lasers have been developed and tested Beam diameters range from 0.6 mm to 2 ram, with under laboratory conditions, but only a few commer divergence angles between 0.4 to 1.5 mrad. Although known as Alpha has been designed to produce 5 MW efficiency is not a primary concern for ground-based of cw power. Lower power Hf and DF lasers are lasers, km lasers achieve overall efficiencies signifi- commercially available, with multiline continuous cantly below 0.01% for multiline operation, and less power levels up to 150 W. Pulsed laser systems are for single line operation. Dye lasers, by contrast, may also available with energies ranging from 2-600 mJ achieve up to a 25% conversion of the pump light with pulse repetition frequencies of 0.5-20 Hz. into laser light. A typical cost for an argon-ion laser Typical beam diameters are 2--40 ram, and beam capable of providing several watts of power is on the divergence angles range from 1-15 mrad. The lasers order of $50,000. require maintenance every 50-100 hours, primarily to change vacuum pump oil and clean the 1-1z or D 2 Neodymium lasers constitute a class of semiconductor injectors. A typical cost for a 150-W commercial DF lasers in which neodymium is used as a dopant in laser is on the order of $90,000. various host materials. The most common neodymium laser is the Nd:YAG, in which a synthetic crystal Carbon dioxide lasers can produce continuous power of yttrium aluminum garnet {YAG) serves as a host levels from milliwatts to several kilowatts at wave- for the neodymium impurity. Alternative host materi- lengths between 9 and I1 microns, with single line als include yttrium lithium fluoride and yttrium operation at 10.6 pm commonly available. The lasers aluminate, although neither enjoys the wide commer- are robust, with lifetimes of several thousands of cial acceptance of Nd:YAG. The lasers are generally hours. Beam diameters range up to several mm. with pumped by flashlamps to produce the required popu- typical beam divergence angles of a few mrad. lation inversions, although diode pumped Nd:YAG Power conversion efficiencies range from 5-15%. and lasers have been operated at sub-Watt power levels. costs for the higher power 5-15 kW CO 2 lasers may Continuous or pulsed powers of several hundred watts run to several hundred thousand dollars. Photovoltaic are available in commercial Nd:YAG lasers at a cells do not respond to radiation at 10.6 microns, wavelength of 1.06 microns. Multiple-rod cavity however, and the CO., laser is thus not suited for designs and laser coupling techniques have been used power beaming to an electric thruster. However. the to achieve Nd:YAG cw-power levels up to a few CO z laser may be useful for concepts which directly kW. z9 Lifetimes are generally limited by the pump absorb the laser energy to heat a propellant. flashlamps to around 10_ shots. Beam diameters range from roughly I mm to l0 ram, with beam Copper val_,r lasers are inherently pulsed lasers divergence angles of a few mrad to tens of mrad. which operate at repetition rates of several kilohertz. Such large divergence angles may present focusing The copper is heated and mixed with neon. producing and collimation problems in beamed power appLica- a vapor which acts as the active lasant medium. A tions. Neodymium lasers have been used for a fast electrical discharge is used to directly excite the variety of applications ranging from military targeting vaporized copper atoms, producing a population to industrial welding, and have proven to be quite inversion and subsequent laser emission. Copper robust under most operating conditions. The cost of vapor lasers emit two lines simultaneously, at 510.6 commercially available high power Nd:YAG laser nm and 578.2 nm, which can be separated in the systems is on the order of $100,000 - $200,000. output beam. Average power levels in commercial units range from a few watts to tens of watts, with Chemical laser_, such as the hydrogen fluoride and overall efficiencies slightly below 1%. Beam diame- deuterium fluoride lasers, use chemical reactions to ters range from 20 to 80 mm, with divergence angles excite a light-emitting species. Hydrogen fluoride of 3-5 mrad. Smaller divergence angles of 0.3-0.5 (HF) emits at 2.6 to 3.3 microns, a region where mrad may be obtained with unstable resonators at the atmospheric absorption is strong. Deuterium fluoride cost of reduced laser power levels. Due to the migra- (DF) lasers operate in the wavelength range of 3.5 to tion of copper from the discharge region, new metal 4.2 microns, where atmospheric transmission is good must be loaded into the dischalge tube after a few but the efficiency is lower and costs are higher due to hundred hours of operation. With copper replenish- the use of deuterium. Large chemical laser facilities nlent, commercial units have expected lifetimes of a have been built for military weapons research. The few thousand hours, l.)epellding upon the power Mid-InfraRed Advanc_v,.I Chemical Laser (MIRACL) level, commercial copper vapor lasers cost from laser is a DF laser which reportedly can produce up $30,000 to $100,000 per unit, with the cost of re- to 2 MW of continuous power, and a large HF laser placement tubes ranging from $600 to $10,000.

6 High power copper vapor lasers have been used for a development are geared more toward FEL user facilities, where researchers may sign up for time on number of years at the Lawrence Livermore National the laser? 2 A compact FEL is being developed for L,aboratory (LLNL) as part of the uranium isotope commercial use at wavelengths from 30-1000 pro?" separation facility. _ The Atomic Vapor Laser Isotope which unforttmately are not suitable for photovoltaic Separation (AVLIS) facility uses a chain of twelve conversion applications. There is no particular copper vapor lasers to pump dye lasers, which in turn conmlercial driver for lower power FEL lasers at are used in the isotope separation process. The wavelengths of interest for ground-to-space power copper vapor lasers produce roughly l0 kW of pulsed beaming, and it is not likely that FEL technology will power with a beam quality roughly 15 times the be available for near-term mission applications. The diffraction limit, with a planned upgrade to 15 kW average power and a beam quality roughly five times potential military and commercial uses for muitimega- the diffiaction lin'fit. LLNL has recently used a watt free electron lasers provide sufficient impetus for the continued development of FELs, which may copper vapor laser to pump a dye laser to produce a synthetic guide star in the atmospheric sodittm layer. _i eventually find additional applications in the high The pulsed output of the dye laser, tuned to 589 rim, power ground-to-space laser beaming applications discussed above. was directed through a 1 meter telescope to produce a fluorescence beacon in the sodium layer. The Based on the brief reviews presented above, Nd:YAG pulsed laser combination produced an incident power of I kW, roughly four times the power required to or copper vapor laser pumped dye lasers may offer the best available technologies for ground-based saturate the sodium layer atoms. LLNL is currently investigating pulse-stretching techniques to increase power beaming to photovoltaic recievers. High laser the return signal intensity. powers can be achieved by coherently combining the output of several individual lasers, and total power Semicondttctor diode lasers have recently been levels of a few kW are probably realistic. The better manufactured which can produce up to 1 Watt of beam quality of copper vapor laser pumped dye lasers continuous power at wavelengths fi'om 770 nm to 840 may give them an advantage over neodymium-based nm. The lasers are made by metal-organic-chemical lasers in terms of collimation and focusing, but the shorter discharge tube lifetimes will require more vapor deposition, with a typical emission region of frequent maintenance and replacement. Coherent only 160 tam x I tam. Advances in fabrication semiconductor diode laser arrays are a promising technology make possible the construction of several thousand emitting regions packed in closely spaced near-term technology for achieving kilowatt power levels at wavelengths of interest for photovoltaic arrays, allowing potentially high power laser opera- conversion, provided issues of phase locking and heat tion. To provide a useful kW--class beam at the receiver, several thousarrl diode lasers would have to removal can be adequately addressed. " be coherently combined and controlled, which is Photovoltaic Receivers. beyond the present capability of current phase locking technology. Other issues include the integration of electrical current controls for each of the laser diodes, Photovoltaic cells illuminated by solar radiation have and sufficient heat removal from packed, heat-sensi- been extensively used to provide electrical power for tive diode arrays. Although a suitable kW-class dit,de a variety of spacecraft applications. Several photovoltaic materials have been developed to convert solar array is not presently available, continued advance- ments in semiconductor laser diode arrays may make illumination to electrical power, as illustrated in this a promising near-term technology for ground-to- Pigure 5. The most mature cell technologies, silicon and gallium arsenide, have solar conversion efficien- space power beaming applications. cies of approximately 18-20% and 23-24%. respec- Kilowatt-class free electron htsers (FELs) are being tively, measured in vacuum (air-mass zero) at a solar intensity of I sun (I.38 kW/m2). The photovoltaic develol_Xl for commercial application, but to date are conversion efficiency can be considerably higher expensive, complex, and not generally available. under monochronmtic illumination, as illustrated in High power near-infrared FELs can be designext and manufactured to suit individual user needs, but at Figure 6. Between 800-1000 nm. the photovoltaic tremendous cost. The component technologies conversion efficiency of silicon cells increases to approximately 40%, double the conversion efficiency typically require large operating areas, and the devices under solar iUumination at comparable intensities. are not particularly robust. Trends for commercial The conversion efficiency of gallium arsenide cells increasesto over 50% at wavelengths between 800- response of 12-18% under solar illumination. Al- 860 nm, again doubling the conversion efficiency though capable of providing better conversion effi- achieved with comparable intensities of solar illumi- ciencies than standard silicon cells under Nd:YAG nation. illumination, CulnSe s cells are still developmental.

The cell efficiencies displayed in Figure 6 drop off With wavelengths between 770-840 nm, semiconduc- fairly linearly at wavelengths below optimum, and fall tor diode lasers are nearly optimum for illuminating rapidly to zero for wavelengths larger than the opti- both silicon and gallium arsenide arrays. Unfortu- mum wavelengths. The necessity to operate at or nately, the technology necessary to produce kilowatts near the optimum wavelengths for efficient power of power from a coherent array of several thousand conversion places additional constraints on the laser semiconductor diode lasers must still be developed. systems which may be used for power beaming. The Deuterium fluoride laser wavelengths, ranging from copper vapor laser, which operates at wavelengths of 3.5-4.2 _m, are too long to be efficiently converted 510.6 and 578.2 nm, does not provide significantly with photovoltaic arrays, even with frequency dou- improved performance for silicon arrays compared to bling. Successive doubling could move the operating the achievable conversion efficiency under solar wavelengths to the correct region for conversion, but illumination. Somewhat better performance is at the cost of significantly reduced laser powers and achieved by gallium arsenide arrays, with an increase added system complexity. in conversion efficiency from approximately 25% under solar illumination to roughly 35-40% under Current solar array designs consist of planar rigid copper vapor laser illumination at the same intensity. panel arrays, flexible fold-out planar arrays, and Using the copper vapor laser to pump a dye laser may concentrator arrays. Rigid panel silicon arrays _7 are provide better efficiencies, as the dye laser can be the most commonly used to date. with panel specific tuned to coincide with the optimum wavelengths for masses of around 23 kg/kW. Total array specific efficient conversion. Copper vapor lasers are inher- masses, including the panel, hinges, booms, harnesses. ently pulsed systems, however, and issues with support structures, power transfer, and launch reten- efficient cell conversion under pulsed laser illumination mountings, are on the order of 35 kg/kW. tion must be addressed. If the pulse repetition period Gallium arsenide rigid panel arrays :7 have a panel is shorter than the minority carrier lifetime of the specific mass of about 19 kg/kW, and a total array photovoltaic cell, the cell responds to the laser illumi- specific mass of around 30 kg/kW. The most ad- nation as if it were essentially a cw laser operating at vanced fold-out array currently being developed is the the average laser power. If the pulse repetition period Advanced Photovoltaic Solar Array (APSA), -_72g is longer than the minority carrier lifetime of the cell, which consists of silicon cells on a kapton blanket. the cell responds to each individual pulse at the peak The specific mass of the APSA array panel is 7.2 laser power. The power lost to series resistance in kg/kW. Using thin film CuinSe 2 cells, the projected the cell increases linearly with peak incident power, specific mass is lowered to around 5 kg/kW. Concen- and the cells must be designed to minimize series trator arrays, which foetus incident light onto a small resistance losses under pulsed illumination? TM area of photovoltaic cells, are designed to increase the intensity and power output over that achievable under Neodymium lasers operate at 1.06 _um, slightly I-sun illumination. Estimated specific masses for beyond the optimum wavelength range for silicon conceptual fresnel lens concentrator panels '7 are on cells and considerably beyond the wavelength range the order of 15 kg/kW. suitable for gallium arsenide cells. At 1.06 !urn, the silicon cell efficiency is approximately 15%, slightly The improved efficiency of photovoltaic arrays under less than the conversion efficiency achieved with solar laser illumination allows the array mass at a given illumination. There is evidence that operating the power level to be reduced. For example, a rigid silicon cells at high temperatures improves the effi- planar silicon array has an efficiency of around 18% ciency at longer wavelengths, "_'_and peak cell re- under solar illumination and a specific power of 35 sponses have been shifted to wavelengths as long as kg/kW. To produce 1 kW of power under solar 1.03 _m. Additional techniques such as light trap- illumination, the total array mass wotdd be on the ping _ might be used to further improve silicon cell order of 35 kg. Because the efficiency of this same conversion efficiencies under Nd:YAG illumination. array is nearly doubled under laser iih,mination of the As noted in Figure 6, CulnS% cells have a peak proper frequency at the same intensity, an array mass response of about 28% at i.06/am, compared to a cell of only 17.5 kg would be required to produce the

8 same 1 kW of power. The mass savings cot,ld result heater powers approaching 500 W. in lower launch costs, or additional propellant could be carried along to increase the operational life of die Kilowatt-class hydrazine arcjets have been developed satellite. for stationkeeping applications? 2 Arcjets heat a propellant via an electrical arc struck between a Low Power Electric Propulsion. cathode and concentric anode, with the anode serving as a nozzle for propellant expansion. The propellant Significant propellant mass reductions or extended temperata_re significantly exceeds the temperature achievable with chemical rockets or resistojets, spacecraft orbital lifetimes can be achieved using electric propulsion, which provides higher specific providing substantially higher values of specific impulse (Isr) than chemical auxiliary propulsion sys- impulse. Figure 8 displays specific impulse versus tems. For a given l,r, the propellant mass (M r) power for various hydrazine decomposition product required to produce a velocity change AV for a mass flow rates, for power levels from 0.3-1.1 kW. _nacecraft with initial mass _ is given by: l,p and thrust values of approximately 450 s and 0.16 N, respectively, were achieved for mass flow rates of M r = Mo 11 - exp[-AV/(g*ln,)] } (k8) (5) 3.73x10 5 kg/s at input power levels approaching 1 kW. where g is the acceleration due to gravity (9.8 m/_). Engines which provide higher l_r values require less A flight qt,alified i.8-kW hydrazine arcjet has been propellant for a given L3V, conserving the propellant developed for north/south stationkeeping duties on the carried into orbit or reducing the launch nmss and Martin Marietta Series 7000geosynchronous commu- cost. nication satellites. =_ The arcjet produces a mission average specific impulse of 520 s, and has demon- Auxiliary chemical propulsion systems for long-term strated an equivalent mission lifetime in excess of 12 satellite stationkeeping applications use monoprope- years. Measured thrust was between 0.15-0.3 N, llant hydrazine thrusters, with a specific impulse of depending upon the propellant mass flow rate. The around 220 s, or nitrogen terroxide/monomethyl 1.8 kW arc jet has been baselined for use on the AT&T Telstar 4 satellite, scheduled for impending hydrazine (NTO/MMH) bipropellant thrusters, with a specific impulse value of approximately 310 s. launch. Developmental work continues on low power Iridium-coated rhenium rockets capable of operating hydrogen arcjets, which can attain specific impulse at high chamber temperatures are being developed, a9 values of 650-1200 s at power levels of 1-4 kW. 4"_ which may provide an increase of tens of seconds in I_ over current chemical auxiliary propulsion systems. Solid propellant pulsed plasma thrusters (PPTs) have Monopropellant hydrazine and NTO/NIMH have been been used on a variety of spacecraft for drag makeup used for satellite maneuvering and stationkeeping and stationkeeping. = A solid fluorinated polymer bar applications. Due to the reduced system complexity, is inserted between two planar electrodes, and an arc is struck across the face of the bar, ablating material. monopropellant hydrazine thrusters have been pre- ferred for small satellite propulsion applications. The material is accelerated via Lorentz forces arising from the interaction of the discharge ct=rrent and an

The performance of monopropellant hydrazine thrust- induced perpendicular magnetic field, with gas- ers, in which the liquid hydrazine propellant is dynamic forces accelerating the remaining ablated decomposed to constituent gases before expansion neutral propellant mass. Specific impulse values through a nozzle, can be augmented by passing the vary from 300-1500 s. Average powers are on the order of tens of watts, with peak powers during decomposed propellant through a heat exchanger resistively heated by an electric current. Resistively discharge approaching a few megawatts. A pair of heated thrusters (resistojet_) have been operated with PPTs are Ct,Tently in use on each of the U.S. Navy a variety of propellants? ° The most prevalent resisto- TIP/NOVA navigation satellites, providing approximately 0.4 mN-s impulse bits per thruster. On the jet systems use hydrazine propellant, with propellant storage and feed systems nearly identical to conven- average, each pair of thrusters fl_e slightly more than tional hydrazine thrusters. The augmented catalytic once per minute to keep the satellites in a precisely defined orbit. 4"_ With a calculated lifetime of 65 thruster (ACT)" has demonstrated long life perfor- million firings per thruster, the initial fuel mass of 1 mance with l,p values approaching 305 s (Figure 7a) with thrust levels of 0.18-0.34 N (Figure 7b) for Ib (0.453 kg) per thruster will last approximately 22 years. interaction of the discharge current with self-induced and/or applied magnetic fields. 4. Additional accelera- Stationao, plasma thnlsters (SPTs) have been devel- tion is provided in applied-field devices by the oped and flown on several Soviet spacecraft. _ The conversion of plasma angular momentum to directed Slrl" sustains an electric discharge between an external linear momentum in the diverging applied magnetic cathode and an anode channel, which ionizes a field. Steady-state MPD thrusters have been operated propellant gas. Ions are accelerated by the channel at power levels from 10-600 kW, and pulsed quasi- electric field, and the plasma is volume neutralized by steady devices have been operated fi'om kilowatts to cathode electrons. SPT's have been operated with megawatts of power. 49 Specific impulse values are xenon propellants at powers up to 700 W, providing typically on the order of a few thousand seconds, with I_r values fiom 1000-2000 s at thrust levels of around thruster efficiencies of 20-30%. Pulsed MPD thrust- 0.03 N. The SPT's have successfully operated for ers flown on the Japanese MS-T4 spacecraft produced several hundred hours in orbit, and for several thou- l,p values around 2500 s with instantaneous efficien- sand hours of ground testingJ _ cies of 22%. The thruster accumulated over 400 firings during 5 hours of operation, successfully bm thn_.wers have been investigated for auxiliary and demonstrating quasi-steady M PD operation in a space primary propulsion for several years. 47 Ions are environment. High power MPD thrusters were also formed in a discharge chamber through collisions with used as plasma sources on the Space Experiment with electrons emitted from a hollow cathode. The ions Particle Accelerator (SEPAC) space shuttle tests, are electrostatic, ally accelerated through a set of which evaluated spacecraft charging effects. charged, perforated ion optics, and the beam is volume neutralized by electrons emitted from a Recent specific impulse values exceeding 5000 s with second hollow cathode placed outside the discharge efficiencies greater than 50% have been reported for chamber. High power mercury ion thrusters operated high power MPD thrusters using lithium ".9and hydro- at power levels of 20-200 kW achieved specific gen '° propellants. Lithium is a condensible propel- impulse values of several throusand seconds. Cunent lant, however, and may not be appropriate for attxilia- research efforts are focused on the development of ry propulsion applications where spacecraft surfaces inert gas ion thrusters operated at 0.5-5.0 kW, which might be coated or contaminated with propellant take advantage of the modest power levels currently backflow. The preliminary performance measure- available for spacecraft applications. ments obtained with hydrogen may have included trace amounts of eroded insulator material, which NASA has taken a low-risk approach to facilitate could surreptitiously contribute to the thrust and raise implementation of low power ion thruster technology the inferred specific impulse. "_' Additional thruster for auxiliary propulsion. 4s In this approach, a 30-cm performance measurements with hydrogen and deute- xenon ion thruster, originally developed for higher rium propellants and a careful examination of erosion power primary propulsion, is instead operated at a products are planned. "_t fraction of its design power level. Figures 9a and 9b show measured specific impulse and thrust values for In deference to near-term space power constraints, a a derated 30-cm ion thruster operated with xenon recent system analysis" _evaluated the potential perfor- propellant at power levels between 0.25-2.0 kWfl mance of high power pulsed M PD thrusters operated The specific impulse varies from a low of 1000 s at at average powers of 10-40 kW. Results of the study 250 W to approximately 2500 s at 2 kW, with a peak indicate that substantial mass savings can be obtained thrust-to-power ratio of 57 raN/kW. Advantages to for LEO-GEO transfer missions with payloads of using a derated thruster for auxiliary propt=lsion 1000-2000 kg, reducing both launch mass and associ- include the elimination of known life limiting issues. ated launch costs. Although pulsed MPD thrusters increased thrust-to-power ratios, and redttced flight are still developmental, the study suggests that pulsed qualification timesfl MPD thrusters may offer significant near-term benefits for orbital transfer applications. Magnetoplasmadynamic (M PD) thrusters use electromagnetic acceleration to achieve high specific im- A variety of other concepts exist which may have far- pulse. An arc struck between two concentric elec- term benefits for primary and attxiliary propulsion. trodes ionizes a neutral propellant gas, which is then The ptdsed inductive thntster (PIT) uses pulsed accelerated by the Lorentz force arising from the electric currents in a flat spiral coil to create transient

10 magnetic fields, which in turn create strong electric photovoltaic materials, such as CulnSe2, or the fields near the coil surface. The electric fields ionize development of kW--class semiconductor diode laser a gas propellant injected over the coil surface, and the arrays, can further enhance the efficient coupling interaction of the coil current with an b_duced plasma between ground-based lasers and spaceborne photo- current accelerates the ionized propellant away from voltaic receivers. Available electric propulsion the coil surface. Specific impulse values of 4000- systems include hydrazh)e resistojets, hydrazine 8000 s at efficiencies exceeding 50% have been arcjets, pulsed plasma thrusters, stationary plasma thrusters, and derated ion engines, with pulsed MPD reported for ammonia propellants. '_ Current PIT thrusters a near-term option. The addition of adaptive devices operate at high average powers, and efforts must be made to evaluate thruster performance at optics, which have the demonstrated capability to average power levels more suitable for near-term propagate nearly diffTaction-limited laser beams through the atmosphere, completes the list of avail- space applications. Other advanced propulsion able technologies required for ground-based laser concepts such as the microwave electrothermal thruster _ (MET) and the helicon wave plasma thnrst- power beaming for space propulsion applications. er,:: are actively being developed in the 10-20 kW Baseline Technologies. power range, and promise efficient operation at specific impulse values of interest. Far-term propul- To facilitate the evaluation of low power ground- sion concepts such as laser sustained plasma thrusters "_and laser thermal thruster._ 7 have been analyzed based laser beanfing applications, the following baseline technologies will be assumed. The ground- and experimentally evaluated at variot, s power levels with a number of propellants. Specific impulse based laser system consists of a coherently coupled values up to a few thousand seconds have been set of copper vapor laser-pumped dye lasers, tuned to achieved with efficient laser power coupling to the an appropriate atmospheric transmission band at a wavelength suitable for photovoltaic conversion by propellant or to a thermal heat exchanger. Although GaAs cells. To minimize the required technology significant technical issues remain to bring each of these advanced concepts to fruition, their continued stretch, the combined laser output power is limited to 5 kW. Based on the laser cost estimates outlined development may one day provide substantial benefits above, the cost ofa 5-kW laser system is on the o_'der for auxiliary and primary space propulsion applica- of $1-2 million, with probable additional yearly tions. maintenance costs of around $100-200 thousand (excluding personnel and other operating costs). The Technology Summary. beams are assumed to be coherently combined to

The match between available laser technology and - create a single, 1-m radius laser beam. Significantly smaller initial beam radii will suffer from large di- photovoltaic cell technology is not perfect, but suffi- ffractive spreading at distances of interest, and signifi- cient overlap exists to suggest that ground-based laser cantly larger beam radii may stress the mechanical beaming can be used with current array teclmologies to achieve conversion efficiencies greater than those ability of the ground based laser optics system to available with solar illumination at similar intensities. follow a rapidly moving satellite in LEO. Higher array efficiencies allow a reduction in the Beam expansion optics are used in concert with a array mass required for a given power level, reducing the total mass of the spacecraft. The use of auxiliary deformable mirror adaptive optic system to propagate electric propulsion reduces the propellant mass a nearly-diffraction limited beam through the atmosphere. A return signal for the adaptive optics is required for a given mission or application, allowing a further reduction in the total spacecraft mass or provided by either a retroreflective array placed on a satellite boom at the correct point-ahead distance, or providing longer operational lifetimes for the same by artificial guide stars produced by laser illumination propellant mass. of the atmosphere, in the latter case. a retroreflector must still be attached to the spacecraft body to Possible laser transmitter-photovoltaic receiver combi- nations include dye lasers, pumped by copper vapor provide correct wavefront tilt inl"ormation for the lasers and tuned to the optimum wavelengths for adaptive optics. either silicon or gallium arsenide cells, or Nd:YAG The cost of a 241-actuator adaptive optics system for lasers operated in concert with light trapping or high a 2.5 m diameler telescope (beam expander) is temperature silicon cells. Near-term improvements in estimated at $3.5 million) _ Additional optics, posi-

11 tion control, satellite tracking systems, ground facili- equations can be combined to determine the laser ties, and other contingency costs could easily add an power necessary to provide a given intensity for an additional $5 million to the initial facility cost. initial beam radius to, laser wavelength A, and beam Including the cost of the 5-kW laser system, the propagation distance Z: estimated initial cost is around $10 million per laser ground site. The lasers will need maintenance and Pt = n (0.61 Z ),)2 I (W) (7) occasional replacement, and even autonomous facilities will require personnel for various station func- tions. Assuming that 15% of the initial site cost per Figure 10 displays a log-log plot of the laser power year will be required for maintenance, etc., an addi- required to maintain an equivalent l-sun intensity of tional $1.5 million per year is budgeted for each 1.38 kW/rn 2 as a function of propagation distance for ground station. Over a ten year station life, the total the baseline beam radius of I m and a laser wave- construction and operating cost is thus estimated to be length of 850 nm, which roughly corresponds to the around $25 million per laser ground site. peak wavelength for efficient conversion by a GaAs array. To provide an equivalent l-sun intensity at Power conversion aboard the spacecraft is assumed to GEO requires a ground-based laser power of 1.5 MW, be provided by GaAs photovoltaic arrays, with a which is considerably beyond the capabilities of the conversion efficiency of 24% under solar illumination near-term lasers considered in this sttdy. A 10-m and 50% under laser illumination at an intensity equal diameter adaptive optic system has been suggested tor to l-sun, or 1.38 kW/m z. A planar rigid panel array laser power beaming to the moon; z4 if such a system is most likely to be used for near-term, low power could be built, the minimum laser power required to applications, "_8eald a conservative estimate of 30 provide a l-sun intensity at GEO would be 60 kW, kg/kW is used for the total array specific mass. A still beyond current laser capabilities. The baseline 5- 500-W hydrazine resistojet, l-kW hydrazine arcjet, kW laser operating at 850 nm would reqt, ire an and 1-kW derated ion thruster were chosen to repre- incredible initial beam diameter of 35 m to provide an sent available auxiliar 3, electric propulsion systems. equivalent l-sun intensity at GEO, and the mammoth adaptive optics required to provide diffraction-limited propagation for such a large beam are beyond the HI. LOW POWER APPLICATIONS capabilities of near-term or envisioned technology. The power requirexl for a given intensity decreases as Given the components for low power ground-to-space a fimction of A2, but the range of useful laser wave- laser transmission and conversion, the following lengths are constrained by atmospheric transmission sections provide a fLrst-order look at some potential and photovoltaic array conversion efficiency consider- mission applications and additional issues which must ations to around 600-900 nm (Figures 4 and 6). be addressed. Reasonable changes in initial beam radius and laser operating wavelength will not significantly impact the Power Beaming Limitations. required laser power, and laser beaming to GEO is not a practical application for the near-term, ground- Many of the low power electric propulsion devices based, low power laser systems considered in this outlined above have been advocated for the north- paper. south starionkeeping of satellites in geosynchronous orbit (GEO), 3.6x107 m above the Earth's surface. Power Beaming to LEO. High power ground-txased laser beaming has been suggested as a means to provide an equivalent I-sun Shown in Figure 10 is a horizontal line delineating intensity to GEO satellite solar arrays during eclipse the baseline 5-kW laser power assumed for this study. periods, eliminating the need for onboard storage At this power level, a diffraction-linlited laser beam batteries. 2t The laser intensity (I) at the receiver is at a wavelength of 850 nm and an initial radius of I given by the ground-based laser power (P1.) divided m can provide an equivalent I-sun intensity out to an by the spot area at tile receiver (r_): orbital altitude of around 2000 kin. If the adaptive optics used to propagate the beam through the atmo- 1 -- PL (w/m2) (6) sphere cannot provide diffraction-limited performance, the maximum distance at which the 5-kW laser can maintain a l-sun intensity can be maintained will be The spot radius is given by Equation i, and the two decreased. Because the intensity is inversely propor-

12 =tonaltothesquare.ofthespolsize (Equation 6). a specific impulse and higher propellant mass flow are factor of two increase in spot size requires a factor of able to deliver higher hlcremental velocity changes four increase in laser power to maintain the same during each illumination period. Fewer illumination intensity at a given distance. The spot size is linearly periods are required for the resistojet than for the proportional to the propagation distance (Equation I), arcjet or derated ion thruster, potentially ='educing the hence a factor of two increase in spot size yields the number of ground-based laser stations. The resistojet same intensity for a given laser power level at half expends more propellant mass during a given illumi- the original propagation distance. For the baselined nation period, however, and the reduced number of power level of 5-kW, diffi'action-limited performance laser ground sites is purchased at the cost of more will generate a 1-sun intensity out to 2000 kin. lfthe rapid propellant depletion. This suggests a trade-off between the number and associated cost of the laser system operates at twice the diffraction limit, the maximum propagation distance to maintain a l-sun ground sites necessary to illuminate a satellite to intensity is cut to 1000 km, still within the range of provide a given delta-V, and the potential commercial most LEO satellite applications. return of the satellite as a function of its extended time in orbit. These issues are evaluat_ in more

Figure 11 shows the maximum possible illumination detail below. time per orbit as a function of orbital altitude for Commercial Benefits. ground station viewing angles of :t:4.V and +60" from zenith. The displayed illumination times assume that the satellite trajectories pass directly over the stations. The benefits of using solar-powered electric propul- The illumination time per orb!t will be reduced by the -sion versus lower-l,¢ auxiliary chemical systems are angle cosine for trajectories which do not pass direct- well documented. _9 The more efficient use of propel- ly overhead. For a satellite orbit of 1000 km and a lant at high specific impulse (Equation 5) allows the conservative viewing angle of +45", the maximum total spacecraft mass to be reduced for a given mis- illumination time is approximately 153 seconds for sion, yielding significant savings in launch vehicle each pass over the laser ground site. At 2000 kin, the costs. A further reduction in mass might be obtained maximum iUun'finatiou time for +45 ° is extended to using laser illumination of the solar arrays to power roughly 325 seconds per site. Longer illumination the electric propulsion systems. The higher conver- times can be obtained for viewing angles of :L60", but sion efficiencies obtained under laser illumination beaming through such low angles in the atmosphere allow the same total power to be generated for a significantly degrade s the compensation ability of the given propulsion system using a smaller photovoitaic adaptive optics. array mass. For a given satellite launch mass, the reduction is solar array mass could allow more Equation 5 can be rearranged to solve for the delta-V propellant to be carded, extending the satellite's provided by expending a propellant mass Nip at a operational life in orbit. For commercial satellites, given l_p: extending the life without hlcreasing the total mass (and associated launch costs) enhances the revenue AV = - (g*I,p) ln[l - M/M.] (m/s) (8) generating capacity of the satellite. The additional revenue must be balanced against the cost of building where M,, is the initial spacecraft mass and g is the and maintaining the laser ground stations for the laser acceleration due to gravity (9.8 m/sZ). The propellant system to be of commercial interest. mass expelled pe_.illumination period is given by the product of the illumination time and the thruster mass Communication satellites generate revenue based on flow rate. Figure 12 shows the maxinmm delta-V the number of transponders they carry. Typical U.S. provided per illumination period for a 5f.X) kg satellite domestic communication satellites Ca1Ty up tO 24 as a function of orbital altitude for the baseline 500- transponders, _,6= with each transponder capable of W resistojet, l-kW arcjet, and I-kW derated ion generating revenues of approximately $1.5 million per thruster. The hydrazine resistojet is assumed to year. '_ Satellite lifetimes are generally 10-12 years. _ provide an Is1,of 300s at a mass flow rate of 1.16xl0 'L with operating costs typically less than $1 million per kg/s. The hydrazine arcjet provides an i=r of 4.50 s at year per satellite. '_z Assuming full transponder usage. a mass flow rate of 3.73xl0 "s kg/s, and the derated a 24-transponder satellite is capable of generating a ion thruster provides an I,v of 1500 s at a xenon mass potential revenue on the order of $35 milliolv'year, or flow rate of 3.74x10 6 kg/s. Because of the limited $350--420 million over the expected lifetime of the illumination times (Figure l 1), devices with lower satellite. Typical satellite construction and launch

13 costs are around $100 million for a 24 transponder 34.2 illumination periods/day, and the derated ion satellite,6j corresponding to a net profit on the order thruster requires I03 illmnination periods/day. of $250-300 million per satellite. Assuming a conservative scenario in which a given Average payload power for 24 transponders and ground site can only see the satellite once per day, the attendant housekeeping duties, battery charging number of illumination periods/day required to requirements, etc., is generally around 500 W. a_'6z provide a given delta-V corresponds to the reqtlired Resistance losses and contingency power requirements number of ground sites. Figure 13 indicates that the may increase the primary power requirenmnts to resistojet, and to a lesser extent the arc jet, can provide around 750 W, which can be supplied by a 22.5 kg fairly significant yearly delta-V's for a reasonable GaAs rigid planar solar array. In addition to the number of ground stations, while the derated ion payload power requirements, the solar array must thruster would require a proliferatod system of ground provide additional power for the auxiliary electric stations to provide yearly delta-V's in excess of a few propulsion system. A 500-W resistojet would require hundred m/s. However, the lower mass flow rates all additional solar array mass of 15 kg, which would associated with the higher specific impulse derated increase to 30 kg for a l-kW arc jet or derated ion ion thruster provide it an advantage in prolonging the tltruster. The potential benefit of laser power beam- satellite ]ife in orbit, and the associated benefit of ing accrues fiom the doubling of the photovoltaic generating additional revenue which could compensate array efficiencies under ilhLmination at the proper for the cost of additional ground stations. In addition. laser wavelength. Ground-based laser power cannot a proliferated number of ground stations could be be continuously supplied to a satellite in LEO, and the used to illuminate more than one satellite per day. array mass required for payload power would still amortizing the cost of the ground stations. have be carried along. However, doubling the efficiency of the array with laser illumination allows the Figure 14 displays the extended satellite life for a 500 array mass required for auxiliary propulsion to be kg satellite in an 800 km polar orbit as a function of halved, from 15 kg to 7.5 kg for the resistojet and yearly delta-V for each of the baselined laser electric from 30 kg to 15 kg for the arcjet and derated ion propulsion systems. Recall that the extended life is thruster. For a given initial satellite mass at launch, achieved by reducing the solar array mass necessary the mass saved by reducing the solar array mass to power the laser electric propulsion system com- could be put toward propellant mass, which could pared to the anay mass required for solar power then be used to extend the life and revenue generating alone, allowing extra propellant mass to be carried capability of the communication satellite. without increasing the total spacecraft mass. "]'he 500-W resistojet reduced the array mass by 7.5 kg. For example, consider a 500 kg communication and this mass provided the 7.5 kg of extra propellant satellite in an 800 km low Earth polar orbit, typical of used to prolong the orbital life compared to a 500 kg proposed constellation satellite masses and altitudes. _s satellite, using the same 500-W resistejet, but pow- At 800 kin, the maximum satellite illumination time ered only with solar radiation. Due to the higher is 120 seconds per laser ground site. Figure 13 shows power levels, the I-kW arcjet and l-kW derated ion the required number of illumination periods per day thn_ster each reduce the solar array mass by 15 kg. as a function of yearly delta-V for each of the base- which allows a 15 kg increase in the propellant mass. lined electric propulsion systems. As expected fiom The higher I_=,ion thruster uses less propellant to Figure 12, the resistojet requires the fewest illumina- achieve a given delta-V, and the extra 15 kg of tion periods to provide a given delta-V, while the propellant provides a longer orbital life extension than high-I,p, low mass flow rate derated ion flu'uster achieved with the arc jet. requires substantially more illumination periods. To provide a yearly delta-V of .sO uv's, which is more It may appear that operating the 500-W resistojet with than adequate for drag makeup at 800 km, _ the less total propellant than the higher power ion or resistojet requires 204 seconds of laser illumination arcjet is an unfair restriction. Although 500-W is each day, or 1.7 illunfination perkxis/day: the arcjet typically the limit for long-life resistojet operation, a requires 3.4 illumination periods/day, and the derated test-bed l-kW hydrazine resistojet has been demon- ion thruster requiles 10.3 illunfination pel'iOd_/'day. st,'ated at an l,v of 310 s and a mass flow ='ate of For a yearly delta-V of 500 nYs, adequate for drag 2.93xl04 kg/s. Because the I,.r is nearly the same as makeup and limited orbital maneuvering, the resistojet the 500-W resistojet, the approximate number of requires 16.5 illuminations/day, the arcjet requires illumination periods/day required to achieve a yearly

14 delta-V can be obtained from Figure 13. For a 500 maintain the 500 kg communication satellite in the kg spacecraft at 800 km, the resistojet must fire for 800 km polar orbit for 10 years, Figure 14 displays approximately 204 s/day, for a total propellant mass the additional lifetime beyond the 10 year period that expenditure of 3.5x!0 2 kg/day. For an extra the laser powered electric propulsion system can keep the satellite in orbit. The total cost of the laser propellant loading of 15 kg, the l-kW resistqiet can extend the satellite life in orbit for an additional 1.2 ground station is thus given by: years, which is only 1.3 times longer than the 500-W resistojet using only 7.5 kg of extra propellant and C_ -- $ I 0M + ( 10+N)*$1.5M/year (9) significantly less than the possible life extensions achieved with the l-kW arcjet and l-kW derated ion where C_ is the total station cost, $10M is the thruster. For a yearly delta-V requirement of 500 assumed construction cost, N is the number of years m/s, the l-kW resistojet with 15 kg of extra the laser illuminated electric propulsion system propellant can extend the satellite life only 26 days, extends the satellite life past the 10 years available compared to 33 days for the 500-W resistojet. The with solar powered electric propulsion, and $1.5 is the use of higher power resistojets, operating at estimated yearly maintenance cost of the station. For concomitantly higher flow rates, does not appear to example, if the laser electric propulsion system significantly extend satellite lifetimes over those extends the satellite life an additional 2 years, the achieved with proven, long-life 500-W resistojet total station cost is estimated to be $28 million. technology. The laser ground station cost must be multiplied by the number of stations required to provide the The right side of Figure 14 displays the additional required yearly delta-V. Assume the arcjet is used to revenue generated by a 500 kg commercial satellite in provide a yearly delta-V of 100 m/s to a 500 kg a 800 km polar orbit due to extended orbital life. The communication satellite in an 800 km polar orbit. satellite is assumed to have 24 transponders, with Figure 13 shows that seven laser ground stations are each operating at only half capacity. As noted above, required, assuming that each ground station each transponder can generate approximately $1.5 illuminates the satellite once per day. Figure 14 million per year at full capacity, hence a satellite with shows that the satellite life is extended for roughly 24 transponders operating at half capacity can 1.3 years, and from Equation 9 the cost of a single generate a potential revenue of $18 million per year. ground station operating for I 1.3 years is around $27 At full capacity, the extra revenue generated per year million dollars. Multiplying this amount by seven would be twice the values shown in tile figure. The operational stations yields a total investment of use of ground-based laser power for electric around $190 million. This total ground station cost propulsion allows the satellite life to be extended past must be covered by the extra commercial satellite the life achieved using solar powered electric revenue for the laser beaming concept to be propulsion alone, generating the additional revenue competitive. For example, using a laser powered shown in the Figure 14. For fairly low delta-V arcjet to provide a yearly delta-V of 100 m/s, Figure requirements, consistent with yearly drag makeup, the 14 shows that a 24 transponder satellite operating at extra revenue can be quite substantial. If the satellite half capacity will generate approximately $24 million is required to perform orbital maneuvers requiring a in extra revenue. For the laser sytem to reach delta-V's of hundreds of meters per second per year, economic breakeven, at least eight such satellites must the extra life in orbit and associated generated each be illuminated by the seven laser ground stations revenue may be inconsequential, and the mission once per day. Hence for delta-V's of interest, might just as well be performed with solar powered multiple satellites must be illuminated by each of the electric propulsion. ground-based laser sites for the concept to be financially competitive with solar electric auxiliary As noted previously, Figure 13 can be used to propulsion. estimate the number of laser ground stations required to provide a given yearly delta-V for each of the Figure 15 displays the estimated number of spacecraft baselined propulsion systems, under the conservative which must be illuminated by the laser ground assumption that a single groond site can only stations, as a function of delta-V per year for each of illuminate a given satellite once per day. The total the baselined electric propulsion systems, for the cost of a laser ground station is given by the system to reach f'mancial breakeven. This figure construction cost and total operating cost. Assuming should be read together with Figure 13, which that a solar powered electric propulsion system can indicates the number of ground stations required to

15 providea givenyearlydelta-Vundertheassumed requirementof I ground site per illumination period. Even with the relatively conservative estimates used Additional Considerations. for commercial satellite revenue generation, the total munber of satellites reqtIired for the system to reach The previous examples assumed a 500 kg satellite financial breakeven falls well below the total nunlber mass in an 800 km orbit. As a consequence of of spacecraft generally proposed to populate Equation 5, if the satellite mass were to be increased, communication satellite constellations in similar the number of ilh|mination periods required to provide orbits. _3 a given delta-V using the same baseline electric propulsion systems would also increase. For a 1000 For reasons noted earlier, the laser illuminated kg satellite, the number of illumination periods/day resistojet requires significantly more satellite targets shown in Figure 13 would double. Under the than either the arcjet or derated ion thruster. Of conservative assmnption that each laser site can only particular interest, Figure 15 shows that the I-kW illuminate the satellite once per day, the required arc jet and l-kW derated ion thruster must both number of laser ground sites would also double. illuminate about the same m,mber of spacecraft for increasing the total ground system cost. If the the laser systems to reach breakeven, even though the satellite mass were kept at 500 kg but the orbital ion propulsion system requires significantly more altitude were reduced to 400 kin, the associated ground stations (Figure 13). For example, Figures 13 decrease in the available illumination time (Figure 11 } and 15 show that to provide a 100 m/s delta-V with would again raise the number of ground stations laser illuminated arcjets requires 7 ground stations, required to maintain a given delta-V at the lower each illuminating 8 satellites once per day. The laser orbit. Longer illumination times are available at illuminated derated ion thrusters require 21 ground higher orbits, and the number of ground stations may stations, each of which must also illuminate 8 be reduced to make the system more economically satellites once per day. The laser illuminated ion competitive. Alternative illumination scenarios, such thrusters provide longer extensions of the satellite as basing the lasers on high altitude aircraft, could lifetimes, allowing the cost of operating the more reduce the need fo, adaptive optics as well as numerous ground stations to be recouped. lengthen the illumination time per satellite. In all such scenarios, the orientation of the solar anays with Locating 7 ground sites such that each site sees 8 respect to the laser ground Or air stations must be satellites once per day may be less difficult than taken into account; the laser won't be of much use if trying to locate 2 l ground sites, each of which must the arrays can't see it. also see 8 satellites once per day. However, the 21 ground sites may be able to target more satellites in Other factors must be taken into account to ascertain a multiple satellite constellation. Either way, the use the utility of ground-based laser power beaming to of the minimum number of ground stations to LEO. Although rigid panel photovoltaic arrays are ilhuninate more than the minimum number of most likely to be flown in the near term, progress is satellites required for breakeven will generate a return continuing on photovoltaic arrays of lower specific on the laser station investment. Detailed satellite mass. The APSA panels discussed earlier are

ground track analyses and Site placement surveys projected to have specific masses of around 7.2 should be undertaken to determine the optimum kg/kW. Doubling the efficiency of l-kW array under number of grotmd stations for a given satellite laser illumination would only decrease the array mass constellation. Probable locations for laser ground by 3.5 kg. The life extensions discussed in the stations operating in concert with polar orbiting previous section, based on 15 kg of extra propellant. satellites are near the Earth's polar regions, where would be reduced by roughly a factor of 4. The extra multiple satellite orbits may be acces_d. Weather revenue generated would be significantly diminished. concerns and limited access may pose additional and the required number of satellites to be illunlinated problems, which must be evaluated. A veritable just for breakeven would increase to the point of cornucopia of illumination scenarios exist, each of making the ground based laser system impractical. which must correlate satellite orbits, delta-V requirements, electric propulsion performance, and Low specific mass batteries capable of powering the laser ground station placement. The parametric electric propulsion systems for the required time exploration of these multiple scenarios is left as a periods would also obviate the utility of a proliferated future exercise. ground-based laser system. Nickel hydrogen batteries

16 withspecificenergiesapproaching32W-hr/kghave half capacity, generating a total of $18 million per beenusedforgeostationarysatelliteapplications,and year per satellite in additional revenue. A solar arecurrentlyundergoingtestingfor LEOsatellite powered electric propulsion satellite was assumed to applications3"_Suchbatteries,orderivativesthereof, have an orbital life of 10 years, which was used as a couldcompetitivelysupplythenecessarypowerfor baseline to gauge the performance of the laser propulsionwithoutresortingto thecomplexityof illuminated satellites. multiplelasergroundsites. The estimated total cost of the laser ground stations included the initial cost of $10 million per facility, IV. CONCLUDING REMARKS plus $15 million per facility in operating and maintenance costs over the baseline 10 year satellite An extensive review of commercial and near-term lifetime and $1.5 million per facility per year for each laser systems, photovoltaic arrays, and adaptive optics year of extended satellite life. The extra revenue indicates that low power ground-based laser generated by extending the satellite lifetimes using illumination of photovoltaic arrays in LEO is laser electric propulsion had to cover the full cost of presently feasible. Approximately 5-kW of power at the laser ground facilities for the laser system to be an equivalent l-sun intensity (I.38 kW/m 2) can be competitive with solar powered electric propulsion. The number of satellites required for the laser electric propagated to an approximate orbital distance of 2000 km using coherently coupled dye lasers and propulsion system to reach breakeven was diffraction-limited adaptive optics. Tile efficiency of substantially less than the typical number of satellites photovoltaic arrays can double under laser proposed for communication constellations, indicating illumination of the proper wavelength, allowing a that the ground-based laser system could be reduction in the solar array mass required for commercially viable. auxiliary electric propulsion. The reduced reray mass allows extra propellant to be calried without A variety of issttes remain to be addressed. Ground increasing the satellite launch mass. The additional track analyses and site location studies should be propellant provides an extended orbital lifetime, which performed to determine optimal satellite orbits and can provide additional revenue for commercial ground station placement. Severe weather at critical satellite owners. site locations could require additional ground sites to be built for redundancy, diminishing the economic

A preliminary analysis was performed to estimate the competitiveness of the system. Anticipated near-term commercial viability of a ground-based laser system developments in low specific mass solar arrays and acting in concert with a 500-W resistojet, a l-kW high energy density batteries for LEO applications arcjet, and a l-kW derated ion thruster. The assumed would diminish any competitive advantage currently specific impulse values for each thruster were 300 s, held by low power, grot,nd-based laser concepts, 45U s, and 1500 s, respectively. The lower l_pdevices obviating the need for such systems to be built. required less illumination time to achieve a given delta-V, but expended more propellant mass than the ACKNOWLEDGEMENTS higher I_r thrusters. Assuming a single ground station could illuminate a satellite once per day, the lower l_p devices required fewer ground stations to achieve a This work was performed at the NASA Lewis Research Center under NASA contract NAS3-25266. given delta-V, but their larger propellant expenditure lessened the extended time in orbit.

REFERENCES A trade study comparing the total cost of the laser ground stations required to provide a given deha-V versus the extra revenue getlerated by extending the _De Young. R. J., Walberg. G. D.. Conway. E. J.. useful life of a number of communication satellite and Jolles. L. W.. A NASA High-Power Slxtce-B_tsed was performed. Using conservative cost values, the Laser Research and.4plications Progrmn, NASA SP- initial cost per ground- based laser facility was 464, May 1983. estimated to be $10 million, with an additional $1.5 million per year per facility in maintenance and 2De Young. R. J. (ed.). Second Beamed Space-Power operating expenses. Each communication satellite Wor'L_'hop, NASA CP-3037, July 1989. was assumed to carry 24 transponders operating at

17 "_Gartrell,C. (ed.),Technoh, gy.' IVorkvhop on Laser Through the Am_osphere, SPIE 1060, 1989, pp. 120- Beamed Power: From Earth to Moon anti Other 128. Application.% proceedings of the February 5, 1991 Workshop. NASA Lewis Research Center, Cleveland, __Walsh, J. L. and Ulrich, P. B., "Thermal Blooming OH. Published by NASA Office of Aeronautics, in the Atmosphere', in Laser Beam Propagation in Exploration, and Technology, Washington D.C., 199 i. the Atmosphere, J. W. Strohbehn (ed.), Springer- Verlag, 1978, pp. 223-320. 4Coomes. E. P., Johnson, B. M., and Widrig, R. D., "Space Power Generation and Distribution", Report I_'ryson, R. K., Principles of Adaptive Optics. PNL-SA-17793, Pacific Northwest Laboratory, Acadenfic Press, 1991. Richland, WA, January 1992. _TZollars, B. G.. "Atmospheric-Turbulence 'ltolloway, P. F. and Garrett, L. B.. "Comparative Compen_tion Experiments Using Synthetic Beacons", Analysis of Space-to-Space Central Power Stations", MIT Lincoln Laboratory Journal, 5 (I), Spring 1992, NASA TP-1955, December 1981. pp. 67-92.

SConway, E. J. and De Young, R. J., "Beamed Laser _Humplu'eys, R. A., Bradley, L. C., and Herrmann. Power for Advanced Space Missions', Space Power, J., "Sodium-Layer Synthetic Beacons for Adaptive 8 (3), 1989. Optics', M/T Lincoln Laboratory Journal, 5 (1), Spring 1992, pp. 45-66. 7Conway, E. J., "Overview of Laser Concepts', Second Beamed Space-Power IVor'L_hop, NASA CP- _*Greenwood, D. P. and Primmerman. C. A., 3037, July 1989, pp. 261-286. "Adaptive Optics Research at Lincoln Laboratory'. MIT Lincoln Laboratory Journal, 5 (l), Spring 1992, gChester, A. N., "Chemical Lasers', in High Power pp. 3-24. Gas Lasers 1975, E. R. Pike (ed.), Inst. of Physics, Bristol and London, 1976, pp. 313-409. 2°Murphy, D. V., "Atmospheric-Turbulence Compensation Experiments Using Cooperative 9De Young. R. J., "Overview and Future Direction Beacons', MH" Lincoln l.abomtory Jtmrnal, 5 (1), for Blackixxty Solar-Pumped Lasers', NASA TM- Spring 1992, pp. 25-44. 100621, August 1988. 2_Landis, G. A., "Space Power by Ground-Based _°De Young, R. J., "Kilowatt Multiple-Path "aFle-Ar Laser Illumination", IEEE AES S.vstems Magazine, Nuclear Pumped Laser", Appl. Phyv. Lett., 38 (5), November 1991, pp. 3-7. March 1981, pp. 297-298. ZZLandis, G. A., Stavnes, M., Oleson, S., and Bozek. "Multiple papers in: Free Electron Lasers, J., "Space Transfer with Ground-Based Laser/Electric Proceedings of the 14th International Free Electron Propulsion", AIAA Paper No. 92-3213, and NASA Laser Conference_ Kob% Japan, .August 23-28, 1992, TM-106060, July 1993. spet:iai edition of Nit(lear htstn_ments and Methods in Physics Research, 331 (1-3), July 1993, full edition. -__Glumb, R. J. and Krier, H.. "Concepts and Status of Laser-Supported Rocket Propulsion', J. Spacecraft _ZZuev, V. E., Laser Beams in the Atmosphere, and Rockets, 21 (1), Feb. 1984, pp. 70-79. Consusitants Bureau, New York, 1982. 2_l,andis. G. A.. "Photovoltaic Receivers for Laser l_Fried, D. L., "Limiting Resolution l,ooking Down beamed Power in Space", J. Prop, lsion aml Power. Through the Atmosphere", J. Opt. Soc. Am., 56 (10), 9 II), Jan.-Feb. 1993, pp. 105-112. October 1966, pp. 1380-1384. 2_Weichel, I1.. Atmospheric Prol_,gatimt _:f 1_t_er n4Karr. T. J., "Atmsopheric Effects on Laser Beams, SPIE Short Course Notes SC44. presented at Propagation', Proceedings of the International Society the OE LASE/90 Conference on Optics, Electro- for Optical Engineering, Nonlinear Optical Beam Optics. and Laser Applications in Science and Maniplch_tion and High Energy Beam Propagation Enginenering, Los Angeles, CA, January 1990.

18 2_i-lecht,J., The 1.aser Guidehool¢, McGraw-Hill Technology 1991, NASA CP-3121, pp. 23-I to 23-6, Book Company. New York, 1986. August 1991.

27Clark, R. S., {ed.), Lasers and Optronics 1990 -_XStella, P. M. and Kurland, R. M., "Development Buying Guide, Gordons Publications, Morris Plains, Testing of the Advanced Photovoltaic Solar Array", in IECEC '91: Proceedings of the 26th Intersociety N J, 1989. Energy Convert'ion Engineering Conference, Boston, 2"Pryor, W. L. (pub.), laser Foct_s World 1993 MA, August 4-9, 1991; vol. 2, pp. 269-274. Buyel:_" Guide, Pennwell Publishing COmpany, Tulsa, -_gByers, D., "Advanced Onboard propulsion Benefits OK, 1993. and Status", NASA TM-103174, 1989. 29 Bransch, H. N. and Kugler, T. R., "Welding and 4°Curran, F. M., Watkins, M. A., Byers. D. C., and Cutting with a I-kW Pulsed Nd:YAG Laser", Proc. Garrison, P. W., "The NASA Low Thrust Propulsion bzt. Conf. on Lasers '90, pp. 325-330. Program". AIAA Paper No. 92-3703. 3°Landis, G. A. and Westerlund. L. H., "Laser Beamed Power: Satellite Demonstration Applications', "McKevitt. F. X.. "Design and Development 1AF-92-0600, presented at the 43rd Congress of the Approach for the Augmented Catalytic Tlu'uster International Astronautical Federation, Washington, (ACT)", AIAA Paper No. 83-1255. D.C., August 28-September 5, 1992. "2Curran, F. M. and Sarmiento, C. J., "l,ow Power

:_Collins, G. P., "Making Stars to See Stars: DOD Arcjet Performance'. AIAA Paper No. 90-2578. and NASA TM-103280, At,gust 1990. Adaptive Optics Work is Declassified", Physics T_May, February 1992, pp. 17-2 I. 4'Sovey, J. S., CreTan, F. M., Haag, T. W., Patterson. M. J., Pencil. E. J.. Rawlin, V. K.. and Sankovic. J. -'2Mross, Michael, Vermont Photonics, Westminster, M., "Development of Arc jet and Ion Propulsion for VT, personal communication, August 1993. Spacecraft Stationkeeping', NASA TM-IU6102, -__Landis, G. A. and Jain, R. K., "Approaches to Solar August 1992. Cell Design for th,lsed Laser Power Receivers'. 4_Myers, R. M., "Electromagnetic Propulsion for presented at the 12th Space Photovoltaics Research and Technology Conference, held at the NASA Lewis Spacecraft", AIAA Paper No. 93-1080, Feb. 1993. Research Center, Cleveland, OH, October 20-22, 4"Ebert, W. L., Kowal, S. J., and Sloan. R. F., 1992. "Operational NOVA Spacecraft Teflon Pulsed Plasma -_Lowe, R. A., Landis, G. A., and Jenkins, P., "The Thruster System", AIAA Paper 89-2497, July 1989. Efficiency of Photovoltaic Cells Exposed to Pulsed "_Brophy, J. R., Barnett, J. W., Sankovic, J. 1'_|., and Laser Light", presented at the 1 lth Space Photovoltaic Research and Technology Conference, held at the Barnhart, D. A., "Performance of the Stationary Plasma Thruster: SPT-100", AIAA Paper No. 92- NASA Lewis Research Center, October 21, 1991. 3155, July 1992. -_:Landis, G. A., Sverdrup Technology, Brook Park, "_TSovey, J. S., }lamley, J. A., patterson, M. J., OH, "Receivers for Laser Power Beaming: Summary Rawlin. V. K.. and Sarver-verhey, T. R.. "Ion of the Workshop at SPRAT-XiI", personal Thruster Development at the NASA Lewis Research communication, July 1993. Center", NASA "I'M 105983. December 1992. -_Landis, G. A., "Thin. Light-Trapping Silicon Solar "*"Patterson. M. J., "Low-lsp Derated ion Thruster Cells for Space", presented at the 20111 IEEE Photovoltaic Specialists Conference. Las Vegas. NV, Operation', NASA TM-105787, August 1992. September 26-30, 1988. "_'_lyers. R. M.. Mantenieks. M. A.. and l,aPointe. :TKraus, R., "IVlass Properties Survey of Solar Array M. R., "M PD Thruster Technology', AI AA Paper No. Technologies', in Space Photovolmic Research aml 91-3568, sept. 1991.

19 _°Miller, G. E. and Kelly, A. J., "Plasma Thruster 6=Sperber, R., "Optimal Lifetimes of Communications Erosion Studies", in MAE 1776.38: Electric Spacetaaft", AIAA Paper No. 93-0950, Feb. 1993. PropuLvuion Laboratory Progress report, Dept. of Mechanical aand Aerospace Engineering, Princeton *_Landis, G. A., "Laser Beamed Power: Satellite University, Princeton, NJ, July-August 1992, pp. Demonstration Applications", 1AF-92-0600, presented A5.I-A5.11. at the 43rd Congress of the international Astronautical Federation, Washington, D. C., August 28-September _Choeiri, E., Departmetn of Mechanical and 5, 1992. Aerospace Engineering, Princeton University, Princeton, NJ; personal communication, 1993. 6_Asker, J. R., "Motorola Proposes 77 Lightsats for Global Mobile Phone Service', Aviation Week & _ZMyers, R. M., Domonkos, M., and Gilland, J. H., Space Technology, July 2, 1990, p. 29. "Low Power Pulsed MPD Thruster System Analysis and Applications", AIAA Paper 93-2391, June 1993. _Palaszewski, B. and Uphoff. C., "Polar Observation Propulsion Requirements', AIAA Paper No. 87-0171, -_Dailey, C. L. and Lovberg, R. H., "the PIT MkV January, 1987. Pulsed Inductive Thruster", NASA CR-191155, July 1993. 6'3Vong, D. W., Hen-in, J., and Stadnik, S. J.. "lnteisat VI Nickel-Hydrogen Battery", in Proc. 21st •"aPower, J. and Sullivan, D., "Preliminary lntersociety Energy Conversion Engineering Investigation of High Power Microwave Plasmas for Conference (voL 3), San Diego, CA, August 25-29. Electrothermal Thruster Use', AIAA Paper No. 93- 1986, pp. 1541-1546. 2106, June 1993. _Landis, G. A., "Photovoltaic Energy Converters", in _"Hooper, E. B., Stallard, B. W., and Makowski, M. Technology Wor "kshopon Laser Beamed Power: Form A., "Whistler Wave Driven Plasma Thruster', Proc. Earth to Moon and Other Applications, Proceedings of the lOth S)wal'n_sinm opz Space Nuclear Power and of the February 5, 1991 Workshop, NASA Lewis Propulsion, M. EI-Genk and M. Hoover (eds0, Research Center, Cleveland. OH. Published by NASA Albuquerque, NM, January 1993, pp. 1419-1424. Office of Aeronautics, Exploration, and Teclmology, Washington D.C., 199l, pp. 1-1 to 1-17. "_6Black, J., Krier, H., Zerkle, D., and Glumb, R. J., "Characterization of Laser-Sustained Plasma Behavior During 10 kW" Laser Thruster Tests', AIAA Paper No. 92-3022, June 1993.

_TGlumb, R. J. and Krier, H., Contim,ms Wave (CW) Thermal Laser Rocket Propulsion: Designing an Operational Thruster, Repot t No. CSI-O 1-88, Combustion Sciences Inc., Champaign, IL, July 1988.

._8Stella, P. M. and Flood. D., "Photovoltaic Options for Solar Electric Propulsion', AIAA Paper 90-2529, July 1990.

"_Curran, F. M., Bennett, G. L., Watkins, M. A., Byers, D. C., Brophy, J. R.. and Sercel, J. C.. "An Overview of the NASA Electric Propulsion Program", IEPC-91-002, presented at the 22nd International Electric Propulsion Conference, Viareggio, Italy, October 14-17, 1991.

_°Agos, L. (ed.), The 1989 World Satel/ite Directory, Phillips Publishing Co., Potomac, MD, 1989.

2O 4 Beacon or Star

Atmospheric Turbulence

Telescope Deformable Mirror

Voltages IMDeformable"

irror Driver_

Fast- 1 Oichroic Mirror Steering Mirror Mirror Drives jr, y Tilts

)ichroic

j Mirror I I ! I I I

eco s*ru%: Sensor _

Figure 1. Adaptive optics system for astronomy. '7

Extendable Boom Corner-Cube Array

550 km

Turbulent .Atmosphere Deformable

50-cm Beam Director

Phases Illuminating " 241 I Retroreflector Laser

Laboratory Trailer

241 CCD Imager Phases

Digital I _ 218x Wavefront 218y Serlsor I Wavefront I Reconstructor IJ _ Gradients

Figure 2. Adaptive optics compensation using satellite retroreflectors for ground-based laser power beaming to spacecraft in LEO? ° 21 y Synthetic_. I,\ 2_1_ }¢ S.oon

Atmospheric Turbulence and

_ _ Thermal Blooming I,<'r Te,.coo,r_ I F . • I/AilU _ Be=co. I .... E"'_t " " I _,e, I ueTormaole .,_ Mirror Aperture-Sharing Element

"_-'_. _, Wavefront _' _ Tilt Sensor f .. ', Corrector ¢ ,. • -- ' ElectronicsC°ntr°' I [Reconstructor]

Figure 3. Adaptive optics compensation using laser generated synthetic beacons for ground-based laser power beaming to spacecraft in LEO. L=

02 HzO ¢Oz "H20 CO= 03 H_O C02 - . CO_ HzO COt 03 II II

100 1 I L | I I | I I 1 I t I 1

"_ 40

2O I ! I IL I t I I l 3 S 7 $ 10 |I 12 13 14 15 W=vd_l_ (microns)

_,¢-- Niat mb=r*d _ Xidal* inlrtro4 )_- FK _frtr4td ,[

Figure 4. Atmospheric transmittance at 0-15 p.m.z'_

22 2E

_ 16 z "-';TW" - / ._/f/ I __', __. 12 '.-.-L._ " .---l..,,--l---; ZlIll t- C.uin...ce2 i 1

o 8 _ _ . < 28O 380 380 400 420 440 460 480 5O0 520

AUGMENTATION HEATER POWER (W)

0 ' I ' 1 ' I : I ' I ' I _ I ' I 1.2 1.6 2.0 2.4 BANDGAP (eV) (7a)

Figure 5. Efficiency of photovoltaicmaterials under solar illumination (air mass zero,I._

(raN) (mlb)

6O • • j • - i 360 8O inP 1W/cm 2 70 o 5O 25 C 3O0 I I I

f- o 40 r./} A-' PI = 165 psiai !i "7-i Z 50

$1 CulnSe 2 -- 200 r. 30 ['- 40 +' '+!- % GaSh Z PI = 100 psi 0

"" 20 lO0 > II1 I "Z 0 36O 38O 40O 42O 44O 46O 48O 50O 52O {.) I 1o i!!_i_ ...... " AUGNfENTATION HEATER POWER (W)

i - , : 0 ' " ) 0.3 0.6 0.9 1.2 1.5 1.8 (7b) SOURCE WAVELENGTH (micron)

Figure 7. Augnlented catalytic thruster (a) specific Figure 6. Conversion efficiency of photovoliaic impulse and tb) thrust vs. heater power for various materials under monochromatic illumination. "z feed pressures. (t

23 4.110 i i

o Dc. 0_3,S r:_ 10 6 D [FFP..ACTI 0 N - LIM rTED BKAM_ /

A • Dc. 0._=4 mm .c2o e., & r_ Dc. 0.Sl mm IA 10 5 ¢ ¢q ,.oo_ 0 ;D r, L,07:':7:oV -B: 10 _

e,, o 4 34O .< 10 3 e., 32O t

o 10 2 --y e_ 2'30 a • A , 6 - 1 • , - • • • . 0.3 " 0_4 0.5 0.E 0.7 0.8 0.9 1.0 1.1 POWER(kW) 10 _ i I 102 10_ 10 4 GEO 105

PROPAGATION DISTANCE (kin) Figure 8. Hydrazine arcjet specific impulse vs. power for various mass flow rates and constrictor d iameters. 42 Figure 10. Diffraction-limited groond-based laser power required to sustain equivalent l-sun intensity vs propagation distance. Initial beam radius = 1 n_ laser wavelength = 850 rim.

2000 | • | i

100

Z '2°f 80 soo E I- 6O

Z -v- 40 .< 20

, ? 0 0 , I , ! 1500 2000 2500 1000 aDO i500 2000 2500 SPECIFIC IMPULSE (s) SPECIFIC IMPULSE (s)

(a) (b)

Figure 9. Derated 30-cm xenon ion thruster (a) specific impulse vs. power and (b) peak thrust as a function of specific impulse/t

24 600 t20 , = 500 kg S/C IN 8OO km LEO ] V|EW_NG AN_ I / )" 100 / 500 • < / ! in.

l [_ RESlS'rOJ-"7 Coo =) i v c 80 400 l / / .... XRCJ.r'T (450 =) T -5/ / [ .... nz._rzDto._(_soo,) £-- / t- Z / *" / -'" 3OO o 60 O °.... F- < < Z .°° / .'° / ..'° 20O m 40 / ..*" ..2 / °." t c 20 100 ¢2

0 r T f , o 500 1000 1500 2000 2500 o • 200 400 600 8o0 1ooo 200

ORBITAL ALTITUDE (kin) MAXIMUMDELTA-V PER YEAR (m/s)

Figure 11. Maximum illumination time per site vs. Figure 13. Maximum # of illumination periods per spacccraf_ orbital altitude for viewing angles of day required to achieve a given delta-V per year for +_45" and +60" from zenith. each of the laser-powered electric propulsion systems assuming an initial spacecraft mass of 500 kg at an assumed orbital altitude of 800 kin.

I0 160 0.25 , , , , L t i I ook, 800kmLEo]

-_'¢ TRA.'¢SPONDEP.S P£R $/C . X .-_ _- 0.20 ..... *RcJ_ (_o s) J . / 8 OPEP.ATING AT 50,% FULL CAPACITY E ...... • "c > 4_ =o g,r2 _< o.15 p 6 105 t.- r. g 7: > _3 ¢., " I t-, C 4 72 \\ ..... ARC JET (45o =) "u Z t=.2 I --R_.SXSTO_ET---- DE.RATED ION(3c0(1500s) =) [.- -% x r_ -... t3 2 36

_°°°.

0.00 t r t ' 0 500 1000 1500 2000 2500 0 100 200 300 400 500 600

ORBITAL ALTITUDE (kin) MAXIMUM DELTA-V PER YEAR (m/s)

Figure 14. 500 kg satellite lifetime extension in Figure 12. Maximt, m delta-V per illumination 800 km orbit and corresponding increase in revenue achieved by the baselined laser-powered electric generated per spacecraft as a function of deita-V. propulsion systems as a function of spacecraft orbital altitude.

25 8O i | I i i Z [500k_sic z_800kmLEOI

-E2 24 TRANSPOND_"RS P£R $/C 0 PEP.AI"L_ G 6O

E- C_ 0

0 ¢0 r-

e-' 2O r,j ,._.../ I-'---_a= (,so,)

50 100 1,50 200 250 300

MA.YIMUM DELTA-V PER YEAR (m/s)

Figure 15. Estimated number of communication satellites which must be illuminated in an 800 km orbit as a function of yearly delta-V requiremet for the ground-based laser system to reach cost breakeven.

NOTES

Form Approved REPORT DOCUMENTATION PAGE OMBNo.0704.0188

Public reportingburden for this collectionof informationis estimatedto average 1 hourper response,includingthe time for reviewinginstructions,searchingexisting data sources, gatheringand maintainingthe data needed, and completingand reviewingthe collectionof information. Send commentsregarding this burden estimate or any other aspect of this collectionof information,includingsuggestionsfor reducingthis burden, to WashingtonHeadquartersServices, Directoratefor InformationOperationsand Reports. 1215 Jefferson Davis Highway, Suite 1204, Arlington,VA 22202-4302, and to the Office of Managementand Budget.Paperwork ReductionProject (0704-0188), Washington, DC 20503.

1. AGENCY USE ONLY (Leave b/ank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED January 1994 Final Contractor Report 4. TITLE AND SUBTITLE 5. FUNDING NUMBERS

Low Power Ground-Based Laser Illumination for Electric Propulsion Applications WU-506-42-31 6. AUTHOR(S) C-NAS3-25266 Michael R. LaPointe and Steven R. Oleson

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION REPORT NUMBER Sverdrup Technology, Inc. Lewis Research Center Group E-8306 2001 Aerospace Parkway Brook Park, Ohio 44142

9. SPONSORING/MONITORINGAGENCYNAME(S)ANDADDRESS(ES) 10. SPONSORING/MONITORING AGENCY REPORT NUMBER National Aeronautics and Space Administration NASA CR-194444 Lewis Research Center IEPC 93-208 Cleveland, Ohio 44135-3191

11. SUPPLEMENTARY NOTES Prepared for the 23rd International Electric Propulsion Conference cosponsored by the AIAA, AIDAA, DGLR, and JSASS, Seattle, Washington, September 13-16, 1993. Project Manager, James S. Sovey, Space Propulsion Technology Division, (216) 433-7454. 12a. DISTRIBUTION/AVAILABILITYSTATEMENT 12b. DISTRIBUTION CODE

Unclassified - Unlimited Subject Categories 20 and 36

13. ABSTRACT (Maximum 200 words)

A preliminary evaluation of low power, ground-based laser powered electric propulsion systems is presented. A review of available and near-term laser, photovoltaic, and adaptive optic systems indicates that approximately 5-kW of ground- based laser power can be delivered at an equivalent 1-sun intensity to an orbit of approximately 2000 km. Laser illumination at the proper wavelength can double photovoltaic array conversion efficiencies compared to efficiencies obtained with solar illumination at the same intensity, allowing a reduction in array mass. The reduced array mass allows extra propellant to be carried with no penalty in total spacecraft mass. The extra propellant mass can extend the satellite life in orbit, allowing additional revenue to be generated. A trade study using realistic cost estimates and conservative ground station viewing capability was performed to estimate the number of communication satellites which must be illuminated to make a proliferated system of laster ground stations economically attractive. The required number of satellites is typically below that of proposed communication satellite constellations, indicating that low power ground-based laser beaming may be commercially viable. However, near-term advances in low specific mass solar arrays and high energy density batteries for LEO applications would render the ground-based laser system impracticable.

14. SUBJECT TERMS 15. NUMBER OF PAGES 28 Laser; Power beaming; Electric propulsion 16. PRICE CODE A03

17. SECURITY CLASSIFICATION :18. SECURITY CLASSIFICATION 19. SECURITY CLASSIRCATION 20. LIMITATION OF ABSTRACT OF REPORT OF THIS PAGE OF ABSTRACT Unclassified Unclassified Unclassified

NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89) Prescribed by ANSI Stct. Z39-18 298-1 O2