Outer planet satellite return missions using in situ propellant productiont

Acta Astronautica Vol. 8, No. 5--6, pp. 511-526, 1981 (RD4-5765/81/050511-16502,00/0 Printed in Great Britain Pergamon Press Ltd.

Outer planet satellite return missions using in situ propellant productiont

R. L. A S H AND V. C U D A JR. Mechanical Engineering and Mechanics Dept., Old Dominion University, Norfolk,

VA 23508, U.S.A.

AND

M. L. S T A N C A T I AND J. C. N I E H O F F Science Applications Inc., Schaumburg, IL, U.S.A.

Abstract--In situ production of oxygen and oxygen with hydrogen for utilization as return propellant from the Galilean satellites has been investigated. Europa has emerged as the preferred landing site because of the availability of water ice and its surface temperature. When oxygen is used with methane transported from Earth, a Europa sample return mission requires 4000 kg less estimated Earth launch mass than a vehicle using space storable propellant. Neither methane nor oxygen require active refrigeration at Europa. When oxygen and hydrogen are both utilized to form the primary sample return propellant, the required processor mass increases, but the estimated Earth launch mass requirement is reduced by an additional 550 kg.

Introduction DATA from the recent Voyager I and Voyager II encounters with Jupiter and its satellites have shown that the Jovian sys tem is even more complex than expected (Stone and Lane, 1979a, 1979b). Jupiter and its satellites are similar in many respects , to the solar system. As a consequence, many questions regarding the format ion and evolution of the solar sys tem may be best addressed by intensive investigation of the Jovian system. Even though Jupiter is the closest sun-like system, it still is sufficiently distant that the number of missions targeted there in the next several decades will be very limited. It is therefore critical that any satellite surface mission be as comprehens ive scientifically as possible. Sample return should be an integral part of an intensive investigation, but the added Ear th launched payload resulting f rom inclusion of a fully fueled sample return vehicle appears to make sample return too costly for a near term mission. The purpose of this work is to show that in situ propellant product ion can reduce Earth launch requirements sufficiently to enable incorporat ion of surface sample return in early satellite missions.

Surface sample return must be an important part of any intensive investigation. Continuing discoveries of new information f rom analyses of lunar samples are indicative of the value of a returned sample. Many of the analytic

tPaper presented at XXXIst Congress of the International Astronautical Federation, Tokyo, Japan, 22-27 September 1980.

511

512 R.L. Ash et al.

instruments employed currently either did not exist or provide much higher resolution than was possible during the American and Russian lunar sample return programs. The tremendously successful Viking mission to Mars lacked instrumentation on its landers capable of resolving many basic questions dealing with water. In general, because of the long lead time between hardware selection and actual landing at the surface, much of the instrumentation which is selected must be very limited compared to state-of-the-art terrestrial instrumentation at the actual time of use. It virtually is certain that some type of "critical" analytic instrument will either be of unacceptable accuracy or unavailable on future surface landers. Returned samples are the only source for continuing analytic investigation of planets and their satellites.

Ash et al. (1978) have shown that in situ production of methane and oxygen at Mars is an attractive alternative to the previously identified surface sample return strategies using solid propellant return vehicles. Their analysis showed how the required waiting time between Mars surface landing and the next Earth return opportunity could be exploited to slowly convert radioisotope thermal energy to chemical propellant mass using Martian carbon dioxide and water as feedstocks. Since the energy density (kJ/kg) available from decay of a radioisotope during a terrestrial year is several hundred times greater than the energy density available from the best chemical propellants, the actual transported propellant mass (radioisotope) is nearly negligible. Of course the equipment required to convert radioisotope thermal energy to useful chemical propellant mass represents a significant mass. Their estimates indicated that the total processor and power generation system mass was on the order of half the mass of propellant produced.

Subsequent studies have shown that production of the oxidizer alone offers significant mass reduction opportunities (Stancati et al. 1980). By transporting fuel from Earth and producing the oxidizer in situ, most of the propellant mass is still removed. That is, if the oxidizer-fuel ratios are 6:1 for hydrogen-oxygen and 3.6:1 for methane-oxygen, the oxidizer mass represents between 78 and 86 per cent of the total propellant mass. Methane is the logical fuel choice in these missions because its liquid storage temperature is near the surface temperature of Mars and other outer solar system bodies.

The Galilean satellites of Jupiter (Io, Europa, Ganymede and Callisto) offer intriguing questions for intensive scientific investigation. Not only have many of the ingredients required for life existed on Europa, at least, but the geologic ages of the satellite surfaces appear to range between very young (Io) and very old (Callisto). The volcanic activity at Io defies analogy with any other object in the solar system. If the unexpected results from the Voyager missions are indicative of the kinds of results that may be found on the satellite surfaces, an interactive mission profile would be very desirable. That is, as questions arise concerning interpretation of results during the surface exploration phase of the mission, it may be desirable to redirect either the course of the surface exploration program or the use of certain scientific instruments. When placed in a 1990-2000 time frame, it is possible to assume robotic devices which have sufficient manipula- tive capability to modify instruments or vehicles to effect significant changes in

Outer planet satellite return missions 513

the mission. The single most limiting quantity will likely be available energy. The most dramatic impact of in situ propellant production on intensive surface scientific investigation is the opportunity it presents for establishing a semi- permanent scientific base which can resupply robotic or rover devices with energy to maintain a protracted investigation of the Jovian system. Thus, surface vehicles could be refueled and reprogrammed to perform multiple missions at the Galilean satellites. That type of scenario can reduce drastically the number of follow on missions required to achieve specific scientific objectives. Not only could the costs be reduced significantly, but also the time between successive investigations could be reduced from on the order of a decade to months.

Many of the design details required to define fully the in situ propellant production system for Galilean satellite sample return must be supplied after further study and experimental evaluation. However, the basic system elements and operational strategy have been defined sufficiently to indicate the potential offered by the approach. Two possible systems appear promising: (1) Production of hydrogen-oxygen propellant from water and (2) Utilization of terrestrially transported methane fuel with oxygen produced in situ from water. Both systems use electrolysis of water and consist of essentially the same equipment. The only difference is the cryogenic refrigeration system required for hydrogen. Furthermore, the preferred satellite for basing both in situ propellant production systems is common.

Case for Europa basing Compelling scientific arguments for choosing one of the Galilean satellites as

the primary staging point do not exist presently. Certainly the intense volcanic activity at Io will be a focal point for many investigations, but the volcanism is precisely why Io is not a realistic location for an extended surface stay. All four satellites maintain fixed Jupiter pointing faces, so that they each have surfaces which are sheltered continuously from the radiation emitted by Jupiter. Con- sequently, other considerations should be employed for selecting Europa, Ganymede or Callisto as the propellant processor base.

Some of the important characteristics of the Galilean satellites are summarized in Table 1. It should be noted also that the maximum solar flux is on the

Table 1. Galilean satellite data

Orbit Mean Surface Escape Maximum Bond Satellite Period* Gravity** Velocity** Temperature § ~ o §

days m/sec 2 km/sec K

Io 1.77 1.44 2.30 141 ± ii 0.56 ± 0.12 §§

Europa 3.55 1.31 2.01 139 ± 12 0.58 ± 0.14

GanymBde 7.15 1.34 2.73 154 Z 6 0.38 ± 0.11

Callisto 16.69 1.03 2.27 167 ± 3 0.13 ± 0.06

* Hubbard (1978)

** Newborn and Gulkis (1973)

§ Johnson (1978)

§§ Hanel and coworkers (1979) have found that the maximum measured infrared temperature wag 125 K with ditu~nal excusions down to 85 K.

514 R.L. Ash et al.

order of 50 W/m 2, which is a significant heat load for cryogenic storage systems. The low acceleration of gravity (0.1-0.15 Earth g) is favorable from propulsive and structural load points of view, but it also results in low forces available for producing pressures required to convert water ice to liquid. Since the critical temperatures of methane and oxygen are 191.1 and 154.8 K, respectively, they can be maintained passively as liquids on all the satellite surfaces but Callisto.

Availability of water must therefore be the overriding consideration. Man- deville et al. (1980) have addressed the water question in their recent study of the evolution of the Galilean satellites. Their results have shown that the surfaces of Ganymede and Europa are covered mostly with water ice. Callisto is not. Europa's surface has spectra which are dominated by water ice. Ganymede does not. That condition is reaffirmed by the photographic data of Smith et al. (1979). Consequently, Europa is the preferred target from the standpoint of water availability.

Europa is also the best location for cryogenic storage of propellant as indicated in Table 1. Not only is its surface temperature lowest, but its short orbit period reduces the solar heating time to about one terrestrial day. Hence, Europa has been selected as basing station for most of the in situ propellant production analyses which follow.

Propellant mass requirements In order to estimate the processor size and power requirements for a

particular mission, it is necessary to estimate the performance requirements for that mission. Those estimates have been addressed previously by Stancati et al. (1980) for some missions, but they are being refined continuously as other missions are defined and alternate propulsive systems and strategies are considered. It is therefore appropriate to discuss briefly the baseline missions currently envisioned and interpret those missions in terms of propellant requirements.

Single satellite sample return (JsSR) Sample returns from Callisto and Ganymede have been discussed previously

by Stancati et al. (1980). Details of the calculations employed and the assumed trajectories are presented there. Gravity assisted trajectories have been assumed and round trip mission times on the order of ten years are typical. Three types of return flight propulsion systems were considered: (1) Hydrogen-oxygen; (2) Methane-oxygen and (3) Space storable. Note that the hydrogen-oxygen system actually uses both hydrogen-oxygen and methane-oxygen bipropellant com- binations. The former is used for ascent from the surface, orbit stabilization and departure from the satellite, with both hydrogen and oxygen produced in situ. The latter is used for subsequent departure from the Jovian system, since storage of hydrogen over the year or so required would entail serious boil-off losses. Only oxygen is produced, with the necessary methane fuel being transported from Earth. Space storable propellant is defined as fluorine-hydrazine propellant with a specific impulse of 3.63 km/sec (370 sec) and is considered the "conventional" alternative to in situ propellant production.


Stancati et al. (1980) translated their calculations into Earth launch mass estimates, in order to compare conventional and in situ propellant production (ISPP) modes. The analysis for Europa has been carried out using the same approach and the Earth launch mass summaries for sample return from each of the three satellites are shown in Fig. 1. It is noted that Europa is also the preferred satellite from the standpoint of minimum propulsion requirements, but the variation between the requirements for ISPP systems is not large. On the other hand, the space-storable systems consistently require 4000-7000 kg more Earth launch mass.

The propellant production requirements for hydrogen and oxygen production, as well as oxygen production alone are summarized in Table 2. Even for the case of simultaneous hydrogen and oxygen production, a larger water mass than

16,000

I I - 1 4 , 0 0 0

. 12,000

I0 ,000

z 8 ,000

_J ,~ 6 ,000

o ~

4,OO0

ISPP-LOX/HYDROGEN

- [ ] ISI

sP

EUROPA GANYMEDE CALLISTO

SAMPLE RETURN TARGET

Fig. 1. Earth launch mass requirements for sample return from the Galilean satellites.

Table 2. Sample return propellant production requirements

Hydrogen and OMygen Production

Water Required Water Processing

Satellite Propellant Required Rate

kg kg kg/day

Europa 1098 1559 3.12

Ganymede 1406 1996 3.99

Callisto 1171 1663 3.33

O~,gen Production Only

Water Methane Oxygen Water Processing

Satellite Required Produced Required Rate

kg kg kg kg/day

Europa 295 1003 1129 2.26

Ganymede 378 1284 1446 2.89

Callisto 314 1069 1204 2.41

516 R . L . Ash et al.

propellant mass is required because a fuel rich oxidizer: fuel ratio is utilized. Water requirements for both options are tabulated, as well as the corresponding water processing rate for 500 days of operation. A two year stay at the satellite surface has been assumed leaving more than a 200 day margin for the processor system.

Satellite hopping An alternative to sample return has been considered in the present study. The

system considered was one which lands at Europa, then after producing sufficient propellant, launches a one-way probe to a different satellite. Since separately targeted probes can be deployed during entry into the Jovian system without ISPP, the question addressed was: What mass penalty would be imposed by ISPP? The probable benefits from utilizing a surface launched hopper are the opportunity for deferred decision on which other satellite to investigate and the opportunity for sequential scientific investigation.

Estimated Earth launch mass requirements were approx. 8000kg for both Europa-Callisto and Europa-Ganymede hopping systems using hydrogen- oxygen propellant. The Europa-Io mission has not yet been examined but should be comparable. ISPP produces a slight (100--300kg) improvement over launch requirements for two directly deployed probes using all space-storable propulsion.

Dual satellite sample return (Js2SR) A very attractive option has been identified recently which can enable sample

return from more than one satellite, without incurring the associated large mass penalties. Essentially two probes are deployed during entry into the Jovian system. One probe is targeted for Europa and carries the hydrogen-oxygen processing equipment. The other probe is deployed at either Ganymede or Callisto and carries an oxygen processor, methane fuel and an ascent vehicle which is capable only of ascent and departure from that satellite. Both probes conduct their scientific investigations for approximately two years. The Gany- mede or Callisto probe, ascends from the surface with departure timed so as to rendezvous with the Earth Return Vehicle (ERV) which is launched previously from Europa. The ERV carries the Europa sample and also receives the second sample at rendezvous before departure for Earth. The basic mission designs proceed as follows:

(1) Launch from Earth 9 December 1991 on a 2÷-type AVEGA trajectory. C3 for such trajectories is about 27 (km/sec) 2. At aphelion of about 2.2 AU, an impulse of 555 m/sec lowers the perihelion to produce crossings of the Earth's orbit. Either of these crossings may be used as a gravity-assist opportunity, the latter one at slightly more than two years flight time being preferable.

(2) Earth reencounter on 22 January 1994 is targeted to produce a gravity- assisted swingby which shapes the Earth-Jupiter transfer. The 1994 Earth- Jupiter opportunity is chosen since it is among the best available opportunities.

(3) Jupiter arrival on 3 September 1996 (V~= 5.829km/sec) with a satellite gravity-assisted capture into a Ganymede resonant orbit. Perijove lies slightly inside Ganymede's orbit.


(4) Satellite "touring--multiple encounters of Ganymede and Callisto--to produce a low approach velocity at Callisto. Touring was originally used as a means of exploring several satellites and regions of the jovian system otherwise inaccessible on a strictly propulsive mission. Here, it is used solely for its trajectory-shaping ability to attain a near-Hohmann transfer between Ganymede and Callisto on the final orbit.

(5) Deploy the Callisto Excursion Module on 2 November 1997, which includes a lander and sample acquisition gear, oxygen processor, ascent vehicle and all required methane fuel. This module weighs about 2200 kg at separation.

(6) Continue satellite tour from Callisto to set/up a low velocity approach at Europa on 1 January 1998, capture into low orbit at Europa and descend to the surface. Collect sample and begin hydrogen-oxygen propellant production (about 1250 kg will be required).

(7) Ascend to low orbit at Europa. Depart July 2000 on tour back to Callisto to pick up second sample. Callisto fly by speed will be near-minimal.

(8) Ascent vehicle from Callisto launches, timing departure to obtain hyper- bolic rendezvous with ERV from Europa. The ERV is the active rendezvous partner.

(9) Transfer Callisto sample to ERV and continue tour to match eventual Jupiter departure conditions.

(10) Depart Jupiter 19 September 2000 (V~ = 6.059 km/sec). Total stay time in the jovian system is a little more than four years.

( l l ) Return to Earth 4 June 2003 (V~ =9.4km/sec) with two samples in a capsule designed for direct atmospheric entry. Elapsed trip time is about eleven and one-half years.

Total Earth launch mass for this mission is estimated to be 10,300kg, or slightly less than would be required for any one-body sample return using conventional space-storable propulsion. The estimated propellant mass requirement for hydrogen-oxygen is 1250 kg which corresponds to a water mass requirement of 1775 kg and, therefore, a water processing rate of 3.6 kg/day. The methane-oxygen propellant requirement at Callisto has been estimated to be 366 kg, of which 275 kg of oxygen are assumed produced by a 150 kg processor system on the surface. That system has not been studied in detail presently.

In summary, the two types of sample return missions discussed here can be accomplished using electrolysis of between 2.3 and 3.6 kg of water per day for 500 days. The work thus far has concentrated primarily on oxygen-hydrogen propellant production systems, because they require the lowest estimated Earth launched masses. Since the methane-oxygen system utilizes only the in-situ production of oxygen and since neither methane nor oxygen require active refrigeration at Europa, that system is much simpler than the hydrogen-oxygen processor.

Propellant processor description The hardware system, from the water collection element to water removal

from electrolytic oxygen is the same for both propeltant processor systems. Only the processing rate varies and it changes only by about 60%. As a consequence,

518 R.L. Ash et al.

the hardware systems can be described simultaneously with indication of estimated variations in size and the addition of hydrogen drying and cryogenic storage for the hydrogen-oxygen processor.

Water collection Considering the conclusions drawn by Pilcher et al. (1972) and Mandeville et

al. (1980), it is very likely that the surface of Europa is pure water ice with perhaps a small amount of dust intermixed. The photographic data of Smith et al. (1979) reinforce their conclusions. However, without further verification, collection of water from pure ice cannot be assumed. The depth of ice required to produce the bright surface observed by Voyagers I and II need only be a few millimeters. Furthermore, the possibility of rocks on or beneath the surface cannot be ignored due to the limited resolution of the photographic data. Hence, the water collection system should be designed to operate reliably in a variety of surface materials. It should be able to tolerate dust and rocks and it should require a minimum amount of energy.

Although a variety of mechanical collection schemes (scraping, drilling, stamping and grinding) have been considered, they are not considered presently the preferred approach. Mechanical collection should require about 25 kJ per kg of ice extracted (Bakakin and Zelenin, 1963), which represents a theoretical power requirement of less than 2W, but susceptibility to rock damage and freezing makes reliability a concern. Thermal collection has been selected here as the preferred mode because of its simplicity and potentially superior reliability. However, thermal extraction also has significant design problems.

Three primary design problems have been examined relating to thermal extraction of ice: (1) Thermal contamination of the environment; (2) High thermal conductivity effects in low temperature ice; and (3) Low allowable operating pressures for water condensation. The thermal contamination problem results from steam released during surface heating and can produce local ambient pressure increases capable of degrading cryogenic insulation as well as produce errosion by ice crystals traveling at high velocities. The high thermal conductivity of ice at temperatures near 100 K (Fletcher, 1970) makes it very difficult to concentrate heating to effect phase change. Low gravitational acceleration means that the propellant processor mass exerts only a very modest force on the surface. Consequently, only a small area can be heated and pressurized for water distillation without resulting in forces exceeding the weight of the vehicle.

A preliminary design of a water collection device is shown schematically in Fig. 2. An insulated cylindrical tube with an outside diameter of approx. 5 cm protrudes into a finned sphere approx. 20cm in diameter. A 100W pu23SO2 ceramic heat source is contained within the cylinder and can be moved up and down in the 20 cm long barrel. The entire assembly is attached to a boom which can be extended 1-2 m and exert a downward force of approx. 50 N.

Water extraction is accomplished by pushing the barrel into the surface material, then lowering the heat source into contact with the material in the barrel. Initially a heat flux in excess of 75 kW/m 2 is produced and approx. 15 cm


200 mm DIA

200 mm

5O

~CONDENSATION SPHERE

s FIN

j ~ --HEAT SOURCE _- ---~ / DRIVE MOTOR

~HEAT SOURCE

j ~ I N S U L A T E D z PENETRATION

TUBE

Fig. 2. Schematic cross sectional view of thermal water collector.

of ice can be melted in 2.5 hr. The presence of foreign material may limit the penetration of the barrel and decrease the depth of ice melted, but the material will act typically as an insulator and thus increase the melting rate of the ice that is present. Consequently, an ice melting rate of approx. 1.8kg/day can be accomplished by each collector. The water vapor is condensed in the sphere and collected in the sump at the bottom of the sphere where it is pumped through the boom into the electrolysis cell.

Three ice collectors have been assumed in this analysis, with a total mass of 21 kg, for production of 3.6 kg of water per day. Grinding, where ice chunks are thrown into a collector bin may be the lowest mass approach, but that system is considered less attractive at this time.

Deep coring opportunities If Europa indeed lacks a solid matrix surface (besides ice), there are many

scientific questions that can be addressed by deep coring through the ice. The power supply required for electrolysis of water and refrigeration of hydrogen provides a potential opportunity for sublimating a one to two meter diameter hole tens of meters deep. If the power supply is thermoelectric, it must reject at least 5 kW of thermal energy per kW of electricity produced. That thermal energy could be rejected into the surface, sublimiting a hemispherical cavity,

520 R.L. Ash et al.

which can be further excavated by moving the power supply deeper into the hole. As the material is sublimated away, the cavity walls could be examined and samples taken.

Ultimately, the situation shown in Fig. 3 would exist because the sublimated ice recondenses on the walls above the power supply. By mounting the power supply in a frame for stability and control, the location of maximum heating could be controlled and the system could move laterally as well as vertically. However, that system assumes greater risks than the much simpler surface processor system described previously.

Electrolysis Electrolysis of water to form hydrogen and oxygen is a well developed

technology. The system which appears to offer highest reliability, coupled with high efficiency and low mass and volume is a solid polymer electrolyte system (see Russell, 1980). Current designs have overall conversion efficiencies of approx. 90%, with cell lives approaching 40,000hr. Heald et al. (1978) have selected solid polymer electrolysis units in their preliminary design of a large propellant production system for low Earth orbit. Their analysis assumed the following scaling laws:

Specific Power 4.47 kW/kg/hr Specific Weight 4.25 kg/kg/hr Specific Volume 0.06 m3/kg/hr

where the reference flow rate is the mass flow rate of water in kg/hr. Their specific weight includes a mass allowance for radiant cooling of the cell.

Propellant drying The assumed operating conditions for the electrolysis cell are 420K,

2.75 MPa. Both oxygen and hydrogen leaving the electrolysis cell will be saturated with water vapor. In the case of hydrogen, the water content can be 65% by mass. Obviously, drying must be utilized and condensation is the most efficient primary mode. By cooling the gases to 275 K in a cold trap, nearly all of the water vapor can be condensed and removed (down to several hundred parts per million). The remaining water removal can be accomplished using ab- sorption/adsorption techniques with regeneration accomplished by heating the drying agent intermittently, outside of the flow lines. Based on the work of Heald et al. (1978), that element should have a mass equal to approx. 10% of the electrolysis cell mass.

The threshhold mass for the electrolysis cell, water cooler and dryer is estimated to be l0 kg for 2.3 kg/day with a mass increase to 12 kg for 3.6 kg/day. Those masses are approx. 20 times greater than the scaling law mentioned previously. The power requirement varies between 440 and 700 W.

Methane-oxygen option The system described to this point represents the entire system required for

oxygen production. Oxygen can be maintained as a liquid with temperatures in


SURFACE SCIENCE AND CRYOGENIC REFRIGERATION S~

THERMOELECTRIC

Fig. 3. Sketch showing how the power supply could be deployed separately to excavate for a deep sample.

excess of 130 K at pressure below the electrolysis cell operating pressure and since the ambient surface temperature at Europa does not exceed 125 K, it can be stored passively. Methane can also be stored passively in the sample return vehicle tanks. Hence, the oxygen processor system would have an estimated mass of less than 50k g, (excluding the power generator). Of course, the methane-oxygen ISPP system lacks the ability to resupply other vehicles since the fuel is not produced, but the system is a very small, very low mass system which is extremely simple.

Hydrogen refrigeration Production and storage of hydrogen fuel offers unique opportunities and

problems. The fuel is simple to produce and is capable of producing very high specific impulse propellant as well as resupplying fuel cell systems for other scientific operations. The very low ambient pressure at Europa (much less than 1 Pa) means that evacuated multi-layer cryogenic insulation systems can be used without vacuum pumping, but the ambient temperature is still too high to permit passive storage of hydrogen.

Based on an estimated heating rate of 0.15 W/m 2 from the surroundings to the hydrogen storage tank (including an allowance for line heat gains), the total liquefaction and re-liquefaction cooling requirements vary between 7 and 10 W

522 R.L. Ash et al.

for production rates between 2.9 and 3.6kg of water per day. That cooling requirement assumes a penalty of 100% normal-to-parahydrogen conversion and that the liquid hydrogen is stored at a pressure of 0.4 MPa (26 K).

Heald et al. (1978) have surveyed the technology available presently for hydrogen liquefaction. Three thermodynamic cycles have been used pre- viouslymStirling, Viulleumier and Brayton. Their data suggest that a 7 W refrigerator would have a mass of about 35 kg and a 10 W refrigerator would have a mass of 50 kg. The corresponding power requirements are 210 W for 2.3 kg/day and 400 W for 3.6 kg/day water processing rates.

Reliability is presently a major consideration for the refrigeration system. Since oxygen and hydrogen can both be used as refrigerant, consideration has been given to defining less sophisticated, but potentially more rugged refrigeration systems. Those analyses have shown that an oxygen refrigerant loop, utilizing Joule-Thomson expansion could be operated between ambient conditions and 56 K, while a turbine expander combined with a Joule-Thomson valve could be operated using hydrogen as the refrigerant, between 56 and 26 K. A sketch of that system is shown in Fig. 4.

The power consumption of the Joule-Thomson system increases by a factor of two to four. However, the mass of that refrigerator appears to be extremely low (on the order of 10 kg) and a trade off between electric generating capacity, reliability and refrigerator mass should be considered after further study.

Electric power generation A variety of electric generating systems are compatible with the ISPP

systems discussed thus far. Estimated power requirements range from about 500 W for oxygen production alone (for a CH4/O2, Europa sample return vehicle) to approximately 1500 W for a hydrogen-oxygen, Js 2 SR dual sample return. The

853 Watts

133 K ~ R A D I A T O R ~ _ ~ . } 2 MPa

~, , ~ _ L 8 , 8 K z ~ J ~J, 745 Watts

0 3 kPal , 0 2 ~ ' ~ 4 MPa ~ 0 . 3 kPa

HYDROGEN . 56K . I I t 5 6 K FROM ' {,/]

ELECTROLYSIS 0.4 MPQ /4 MPa , 50 K H_I" '"'

!

IVENTED 4 MP ~ ..~ 0 4 MPo HYDROGEN 44 K IO Mr'a •

EXPANDER STORAGE TANK

Fig. 4. Schematic of thermodynamic cycle utilizing oxygen and hydrogen as refri- gerants.


radioisotope thermoelectric generator (RTG) systems used on the Voyager I and Voyager II spacecraft could be used presently for the oxygen system, but the required mass would be 125 kg (Jakubowski, 1980). That mass is already more than three times larger than the estimated processing system mass it supplies and a 1500W system of that type would be prohibitively large. Four types of advanced power generating systems may be available in the U.S. by the time a Galilean sample return mission occurs: Rankine cycle, Brayton cycle, Advanced RTG's and Small Reactors.

Since dynamic power cycles require separately deployed radiators, that mass is included in generator mass estimates. Figure 5 is indicative of the range of power generators which may be available for a 1990-2000 launch and the mass they represent. The Voyager RTG has been proven in space while the other systems have not. Organic Rankine cycle (Broune et al., 1976), Brayton isotope power systems (Kenney et al., 1976) and advanced RTG systems (Lieberman et al., 1976) all appear to have similar mass characteristics. Both of those dynamic cycles have undergone extensive development programs, but neither has been demonstrated in space. The Stirling cycle system is only at the conceptual stage (Jakubowski, 1980) and therefore may have more optimistic performance estimates than the other dynamic cycles. Minireactor systems (Koenig et al., 1977) appear to offer extremely good performance for power generation levels above 3 kWe, but a 300 kg minimum mass limits their low level power application.

A nominal specific mass of 140kg/kWe has been assumed in this study. Barthelemy et al. (1979) have projected RTG systems with specific mass levels of 130kg/kWe in 1983 and l l5 kg/kW¢ in 1985, so the value assumed may be conservative.

600

500

I 4 0 0 ¢t)

:~ 3OO nr 0 I-- '~ 200 nr W Z hi o I00

VOYAGER RTG - - - - - - ADVANCED RTG - - - - - O R G A N I C RANKINE . . . . B R A Y T O N / . . . . . S T I R L I N G / . . . . . REACTOR S Y S T E M ~ / /

/ . / / /

/ / S - J . i -'--- / . J

/ / ; , -<,-J _ / /~..>'" j j - "

f J "

, I , I , I

I 0 0 0 2 0 0 0 : 5 0 0 0

E L E C T R I C P O W E R - Watts

Fig. 5. Variation of estimated power generator mass with electric power produced for ISPP generator candidates.

524 R.L. Ash et al.

System summary A schematic diagram of the ISSP processor system is shown in Fig. 6. The

hydrogen refrigeration system is not shown because it simply represents a system mass and power requirement at this time. A breakdown of elemental mass, energy and volume estimates are given in Table 3. Even though the smallest methane-oxygen systems may be somewhat smaller than the system size range listed in the table, the more conservative data can be tolerated.

The processor mass requirement of l l0-180kg for Europa sample return using methane-oxygen and hydrogen-oxygen can be compared more readily if the required mass of terrestrial methane (295 kg) is included with the processor

qy_k N I ~

EC, COLAT! : J I O o 01 o o

ELE%I~OL'5 5 1

@ ~o:x "~ ~t : : ? : I I ~ CHECK [ / k J L / ~ . ' - . ~ J CONTROL LJ PUMP VALVE FEED DEIONIZER VALVE

WATER PUMP COLLECTOR

Fig. 6. Flow diagram for ISPP processor (hydrogen is vented prior to sorption for O2/CH4 systems).

Table 3. System summary

Water Collection I~ and valves Water treatment Electrolysis Cell Dryer Radiator Piping and Structure

Hydrogen Refrigerator

POWER Rm~UIR~TS

J SR S

Js2SR SYSTEM MASS

Js SR Js2SR

(Flow Rates: 2.9 to 4.B kg o£ H20/day

Mass Power Voltrae

14-21 kg i0 to 15 l 5 kg 10-20 W 1 5 kg i0 W 51

10-15 kg 440-700 W 7-12 l 1-2 kg 10 W 1-2 l 1-2 kg (1-2 m 2) 4-5 k~ - 1-2 l

40-55 kg 450-740 W 27-37~

35-50 kg 210-400 W 100-250 £

C~ 4 +(02) +(H2/02)

450 W 660 W

740 W 1200 W

ii0 kg 180 kg

190 kg 300 kg


mass. Then, the return propellant mass requirement of 1298 kg is substituted by a landed mass of 405 kg of methane and processor system. The hydrogen- oxygen system requires a landed mass of 180 kg to produce a propellant mass of 1308 kg. The dual sample return missions (Js 2 SR) have similar mass ratios but are less well defined.

A 450kg processor system mass has been used in the initial trajectory analyses and propulsion calculations. That mass includes the terrestrial methane mass for the oxygen production analyses. As the data from Table 3 show, the 450 kg assumption provides a mass margin for all cases except the two satellite methane/oxygen case. The corresponding Earth launch mass requirements have been presented in Fig. 1 for single satellite sample return. A break down of estimated dry landed mass at Europa is given in Table 4.

Discussion and conclusions This analysis has shown that estimated Earth launch mass requirements for a

Galilean satellite sample return mission can be reduced by more than 40% of the requirements for the conventional space storable option when in situ propellant production is utilized. The simplicity of the system utilizing methane-oxygen propellant must be weighed against the ability of the hydrogen-oxygen system to refuel other vehicles, as well as provide an additional five per cent reduction in mass.

Use of methane fuel with oxygen results in a system which cannot replenish its fuel, but the oxygen production system requires virtually no new technology. The RTG's used currently on Voyager I and Voyager II could be employed with existing solid polymer electrolysis cells to build a system with very few moving parts and a total nominal mass which is within the range assumed currently. Questions still remain concerning how best to extract water from a very cold, icy surface, but thermal extraction looks attractive over a wide range of surface conditions.

Hydrogen and oxygen production involves greater risks, but its ability to resupply other vehicles provides that processor system with the opportunity to maintain a semi-permanent base at Europa. Advances in microprocessor and robotic technology by the end of this century may enable robotic devices to repair and maintain the more complex system required. Furthermore, the additional electric power and cryogenic refrigeration capability offer opportunities for more complex scientific instrumentation and experiments.

Table 4. Estimated Europa sample return landed mass

El~_nt Mass

Lander 300 kg ISPP systs~ 450 kg Tanks 3O5 kg 02/H 2 ~ngine 75 kg 02/f~ 4 ~ngine 75 kg Earth Entry Capsule 60 kg Earth Return Vehicle 200 k~

Dry Landed Mass 1465 kg

526 R.L. Ash et al.

The oppor tun i t i es offered by I S P P sys tems at the Gal i lean satell i tes are very a t t ract ive . The sys tem mass requi red for sur face sample r e tu rn is r educed sufficiently to enable its cons ide ra t ion for near term miss ions . Conve r se ly , for the same es t imated Ear th l aunch mass that is requi red for sample r e tu rn f rom one satell i te us ing space s torable propel lan t , I S P P sys tems are capable of enab l ing s imu l t aneous sample r e tu rn f rom two satell i tes.

A ckno wledgements--This work was supported by Jet Propulsion Laboratory of the California Institute of Technology under Contract 955731, which is a sub-contract under NASA contract NAS7-100. The authors would like to thank Warren L. Dowler of Jet Propulsion Laboratory for his helpful and stimulating discussions.

References Ash R. L., Dowler W. L. and Varsi G. (1978) Feasibility of rocket propellant production at Mars.

Acta Astronautica 5, 705-724. Bakakin V. P. and Zelenin A. N. (1963) Excavation of frozen soils. International Permafrost

Conference, Purdue University. National Academy of Sciences, National Research Council Publication No. 1287.

Barthelemy R. R., Koester R. J., Stearns J. W. and Stofel E. (1979) Aerospace power systems--a building surge. Astronautics and Aeronautics, Feb. 1979, 64-72.

Brouns R. C., Krueger E. C., McKenna R. F., Niggerman R. E. and Russo F. A. (1976) KIPS kilowatt isotope power systems. Proc. I l th IECEC Meeting, State Line, Nevada, pp. 1457-1461.

Fletcher N. H. (1970) The Chemical Physics of Ice, p. 144. Cambridge at the University Press Cambridge.

Hanel R. et al. (1979) Infrared observations of the Jovian System from Voyager II. Science 206, 952-956.

Heald D. A. et al. (1978) Orbital propellant handling and storage systems for large space programs. NASA Contractors Report JSC-13967, prepared for Johnson Space Center, Houston, Texas.

Hubbard W. B. (1978) Possible flyby measurements of Galilean satellite interior structure. Icarus 33, 336-341.

Jakubowski A. K. (1980) Radioisotope-powered free-piston Stifling engine for space applications. Acta Astronautica 7, 169-181.

Johnson T. V. (1978) The Galilean satellites of Jupiter: "Four Worlds" Ann. Rev. Earth and Planetary Sci. 6, 93-125.

Kenney W. D., Longee H. W., Adland D. L. and Hemler R. J. (1976) Brayton isotope power system ground demonstrator. Proc. 1 l th IECEC Meeting, State Line, Nevada, pp. 201-208.

Koenig D. R., Ranken W. A. and Salmi E. W. (1977) Heat-pipe reactors for space power applications. J. Energy 1,237-243.

Lieberman A. R., Osmeyer W. E. and Hammel T. E. (1976) The low cost high performance generator (LCHPG). Proc. l l t h IECEC Meeting, State Line, Nevada, pp. 1591-1598.

Mandeville J. C., Geake J. E. and Dollfus A. (1980). Reflectance polarimetry of callisto and the evolution of the Galilean Satellites. Icarus 41,343-355.

Newburn R. L. and Gulkis S. (1973) A survey of the outer planet Jupiter, Saturn, Uranus, Neptune, Pluto and their satellites. Space Sci. Rev. 3, 179-271.

Pilcher C. B., Ridgway S. T. and McCord T. B. (1972) Galilean satellites: Identification of water frost. Science 178, 1087-1089.

Russell J. H. (1980) An update of solid polymer electrolyte water electrolysis programs at General Electric. Technical paper presented at the 3rd World Hydrogen Energy Conf. Tokyo, Japan, 23-26 June 1980.

Smith B. A. et al. (1979) The Galilean satellites and Jupiter: Voyager II imaging science results. Science 206, 927-950.

Stancati M. L. et al. (1980) In situ propellant production for improved sample return mission performance. Astrodynamics, Adv. Astronautical Sci. 40, 909-921.

Stone E. C. and Lane A. L. (1979a, b) Voyager I and II encounter with the Jovian system. Science 204, 945-948; 206, 925-927.

Outer planet satellite return missions using in situ propellant productiont

Documents