NASA/CR—2020-220368

Small Lunar Base Camp and In Situ Resource

Utilization Oxygen Production Facility
Power System Comparison

Anthony J. Colozza
Vantage Partners, LLC, Brook Park, Ohio

March 2020
 NASA STI Program . . . in Profile

Since its founding, NASA has been dedicated • CONTRACTOR REPORT. Scientific and
to the advancement of aeronautics and space science. technical findings by NASA-sponsored
The NASA Scientific and Technical Information (STI) contractors and grantees.
Program plays a key part in helping NASA maintain
this important role. • CONFERENCE PUBLICATION. Collected
papers from scientific and technical
conferences, symposia, seminars, or other
The NASA STI Program operates under the auspices
meetings sponsored or co-sponsored by NASA.
of the Agency Chief Information Officer. It collects,
organizes, provides for archiving, and disseminates • SPECIAL PUBLICATION. Scientific,
NASA’s STI. The NASA STI Program provides access technical, or historical information from
to the NASA Technical Report Server—Registered NASA programs, projects, and missions, often
(NTRS Reg) and NASA Technical Report Server— concerned with subjects having substantial
Public (NTRS) thus providing one of the largest public interest.
collections of aeronautical and space science STI in
the world. Results are published in both non-NASA • TECHNICAL TRANSLATION. English-
channels and by NASA in the NASA STI Report language translations of foreign scientific and
Series, which includes the following report types: technical material pertinent to NASA’s mission.

• TECHNICAL PUBLICATION. Reports of For more information about the NASA STI
completed research or a major significant phase program, see the following:
of research that present the results of NASA
programs and include extensive data or theoretical • Access the NASA STI program home page at
analysis. Includes compilations of significant http://www.sti.nasa.gov
scientific and technical data and information
deemed to be of continuing reference value. • E-mail your question to [email protected]
NASA counter-part of peer-reviewed formal
• Fax your question to the NASA STI
professional papers, but has less stringent
Information Desk at 757-864-6500
limitations on manuscript length and extent of
graphic presentations.
• Telephone the NASA STI Information Desk at
757-864-9658
• TECHNICAL MEMORANDUM. Scientific
and technical findings that are preliminary or of • Write to:
specialized interest, e.g., “quick-release” reports, NASA STI Program
working papers, and bibliographies that contain Mail Stop 148
minimal annotation. Does not contain extensive NASA Langley Research Center
analysis. Hampton, VA 23681-2199
NASA/CR—2020-220368

Small Lunar Base Camp and In Situ Resource

Utilization Oxygen Production Facility
Power System Comparison

Anthony J. Colozza
Vantage Partners, LLC, Brook Park, Ohio

Prepared under Contract NCC12BA01B

National Aeronautics and

Space Administration

Glenn Research Center

Cleveland, Ohio 44135

March 2020
 Acknowledgments

I would like to acknowledge Ian Jakupca and Bill Bennet for their technical review
and Christina Gerber for her editorial review.

Trade names and trademarks are used in this report for identification
only. Their usage does not constitute an official endorsement,
either expressed or implied, by the National Aeronautics and
Space Administration.

Level of Review: This material has been technically reviewed by expert reviewer(s).

Available from
NASA STI Program National Technical Information Service
Mail Stop 148 5285 Port Royal Road
NASA Langley Research Center Springfield, VA 22161
Hampton, VA 23681-2199 703-605-6000

This report is available in electronic form at http://www.sti.nasa.gov/ and http://ntrs.nasa.gov/

 Contents
Summary ....................................................................................................................................................... 1
1.0 Introduction........................................................................................................................................... 1
2.0 Lunar Environment ............................................................................................................................... 3
2.1 Surface Characteristics................................................................................................................ 5
2.2 Incident Solar Radiation.............................................................................................................. 7
3.0 Oxygen Production ............................................................................................................................... 8
3.1 Oxygen Production System Power Analysis............................................................................. 10
3.2 Oxygen Production System Power Requirements .................................................................... 15
3.3 Oxygen Production System Mass Analysis .............................................................................. 18
4.0 Habitat and Base Camp....................................................................................................................... 23
4.1 Environmental Control and Life Support.................................................................................. 23
4.2 Communications ....................................................................................................................... 26
4.3 Operational Equipment ............................................................................................................. 30
4.4 Base Camp Total Power Requirement ...................................................................................... 32
5.0 Power System Sizing .......................................................................................................................... 32
5.1 Photovoltaic Array Power System ............................................................................................ 33
5.2 Nuclear Kilopower Reactor Power System............................................................................... 49
6.0 Results................................................................................................................................................. 59
6.1 Case 1—Power System Comparison for ISRU Oxygen Production Plant ............................... 59
6.2 Case 2—Base Camp’s Power System....................................................................................... 68
6.3 Case 3—Oxygen Production and Base Camp Combined Power System ................................. 72
7.0 Conclusion .......................................................................................................................................... 78
Appendix—Nomenclature .......................................................................................................................... 81
Symbols ............................................................................................................................................. 81
References ................................................................................................................................................... 89

NASA/CR—2020-220368 iii
 Small Lunar Base Camp and In Situ Resource Utilization Oxygen
Production Facility Power System Comparison
Anthony J. Colozza
Vantage Partners, LLC
Brook Park, Ohio 44142

Summary
This report examines the power requirements for operating an in situ resource utilization (ISRU) oxygen
production system on the lunar surface and a small six-person base camp. The baseline ISRU system
produced 1.63 kg/h for a total day and night production rate of 1,154 kg. It was estimated that this plant
would require 25.83 kW of power to operate. The base camp power includes auxiliary equipment as well as
a communications system. The required power estimate for the base camp was 28.05 kW. This estimation
was used to size a power system and determine its mass for meeting these requirements. Three types of
power systems were considered: a solar photovoltaic (PV) array system using batteries for energy storage, a
PV array system using a regenerative fuel cell (RFC) for energy storage, and a modular 10-kW electrical
output power Kilopower reactor system. Three separate cases were examined: a stand-alone ISRU oxygen
production system, a base camp, and a combined ISRU oxygen production system and base camp.
For the PV array-based system, the RFC energy storage method had a mass advantage over a battery-
based energy storage system. For higher power nighttime power operation for all three cases, the RFC
system’s specific energy was just over 830 Wh/kg. For the lower power nighttime “keep-alive” level used as
part of the Case 1 analysis, the specific energy for the RFC was 456 Wh/kg. Both of these levels are
significantly above the specific energy of 200 Wh/kg for the battery. Because of this higher specific energy,
the RFC-based system provided significant mass advantages over the battery-based energy storage system.
The baseline reactor system utilized shielding and separation distance to meet the desired maximum
radiation dose level of 5 rem/yr for personnel operating within the vicinity of the power loads, base camp,
and oxygen production facility. There are methods that could potentially be utilized to reduce the
shielding requirements and separation distance. Implementing these would reduce the overall system
mass for the reactor. Also, optimizing the reactor output to a specific mission would provide benefits in
mass at the expense of modularity.
The results of the power system comparison between a solar PV array-based system and a Kilopower
reactor-based system has shown that for missions required to operate throughout the lunar night at power
levels comparable to those used during the day, the reactor-based system provides a significant mass
advantage. However, for applications that can meet their mission requirements while only having to
operate during the daytime with minimal power required to survive the nighttime, the PV array-based
system provides a mass advantage.

1.0 Introduction
The lunar surface provides a number of challenges for both manned and unmanned missions. A key
aspect to successfully operating on the lunar surface is the availability of a reliable power source that can
meet the mission requirements for the duration of the mission. The power system requirements will vary
considerably depending upon the type of mission (manned or robotic) as well as the individual power
requirements of the components or devices associated with that particular mission or activity. The
estimated power levels for some of the potential missions, devices, and vehicles are listed in Table 1. The
items highlighted in gray were the missions selected for study in this report.

NASA/CR—2020-220368 1
 TABLE 1.—POWER LEVEL RANGE OF VARIOUS MISSIONS OR ACTIVITIES ON LUNAR SURFACE
Mission Power level range, kW Notes
Habitat/base camp power 30.0 to ≥ 100.0 This power level is directly related to size of habitat and
(avg. 5 to 10 kW/person) number of people it will need to support.
Communications system power 0.3 to 1.0 per transmitter The power levels are for powering a single transmitter.
(surface to surface and The total communications system may also be powered
surface to orbit) from a central power source.
Support equipment power 2.0 to 10.0 This power level depends on activity that vehicle is
(vehicles, rovers, etc.) supporting, and whether it is manned or unmanned.
Long range rover power/heavy 7.0 to 30.0 For long-range pressurized rover, the power level will
equipment vehicle depend on the number of people it needs to support.
Charging station power 1.0 to 10.0 The charging station output power will vary depending
on the number of vehicles charged and the desired charge
time.
Electrostatic/electric field 1.0 to ≥ 10.0 A number of factors determine the power level such as
radiation shielding power distance of the field and shielding required. The shield
(cosmic rays and solar protons) uses very high voltage to produce deflection of particles
but not much power.
In situ resource utilization (ISRU) 10.0s to 100.0s (thermal) Thermal and electrical power requirements are highly
processing heat and power (O2 10.0s to 100.0s (electrical) variable depending on process used and rate of
and raw material production) production.
Long distance rover (magnetic 0.1 to 5.0 Power level depends on rover size and type of science
field mapping, general science) performed.
power
Water ice exploration and 10.0s to 100.0s (thermal) As with general ISRU, this will depend on amount of
recovery heat and power 10.0s to 100.0s (electrical) water that is to be processed.
Remote, stand-alone geology 0.1s to 1.0 Sensor requirements, heating for nighttime operation, and
station power communications will determine power requirement.
Remote, stand-alone astronomy 0.1s to 1.0 Communications and nighttime operation are main power
observatory power consumption items.

An in situ resource utilization (ISRU) oxygen production facility and a small lunar base camp were
selected from the missions and activities listed in Table 1 for evaluation. There are a number of different
types of power sources that can potentially meet the requirements of these selected mission components
on the lunar surface. For this analysis, two types of power systems were selected and sized to meet their
requirements; a constant power nuclear system based on the Kilopower technology and a photovoltaic-
(PV-) based system utilizing either batteries or a regenerative fuel cell (RFC) for energy storage.
For comparing the power system to these mission aspects, the following assumptions were made:

• The power systems would provide power for operation over 1 day and night period. The
identified systems would operate either continuously over this time period or during the daytime,
approximately 354 h, and have adequate power to enable the system to survive the nighttime, also
approximately 354 h.
• The ISRU production goal would be to produce at least 1,000 kg of oxygen over 1 day and night
period.
• The habitat and base camp goal would be to meet the power requirements for sustaining six
astronauts over a day and night period.
• The oxygen production and base camp would be located at 30° N latitude.

NASA/CR—2020-220368 2
 Figure 1.—Moon (Ref. 1).

2.0 Lunar Environment

The Moon (Figure 1) is a dry, crater-filled land with no appreciable atmosphere. The temperature
difference from the sunlit areas to shadowed areas can exceed 300 K. The surface is covered with a fine
powdered material called regolith made from eons of bombardment by meteors, asteroids, and comets.
The successful design and operation of a lunar oxygen production system within this harsh environment
will require the system to leverage the advantages of the environment (such as abundant solar energy) and
minimize the disadvantages (such as temperature extremes). From the solar availability to the amount of
removable oxygen in the soil, the capabilities of this system will be based on how well it is adapted to
these environmental conditions.
The Moon orbits Earth every 27.3 Earth days. It is tidally locked to the Earth, therefore, it rotates at
the same rate it orbits Earth. This causes the same side of the Moon to face Earth all of the time. Having
the same side continually face Earth provides an advantage when it comes to communications. A device
or system placed on the Earth-facing side of the Moon can be in constant communication with Earth
without the need for a lunar satellite communications system. However, the day length on the Moon is
29.5 Earth days. This is the synodic period, the time it takes for the Sun to move across the sky and return
to its starting point for a fixed point on the lunar surface.
Another unique orbital characteristic is its low declination angle of approximately 1.5°. This angle
means that there is very little variation in solar elevation throughout the year and little seasonal effect.
Because of this, locations at high elevations near the poles could be in constant sunlight throughout the
year, pending local geographical obstructions. The Clementine spacecraft identified just such a location in
1994. This continuously illuminated location is on the rim of the Peary crater at the Moon’s north pole
(Ref. 2). Some additional physical characteristics of the Moon are listed in Table 2.
The temperature variations between the day and nighttime on the lunar surface are extreme. The
hottest location on the lunar surface is during the daytime near the equator where the surface temperature
can reach approximately 112 °C (385 K), whereas, that same location at night would drop to
approximately –173 °C (100 K). Toward the poles, the daytime temperature does not get very high due to
the low sun angles at these latitudes. The maximum surface temperature achieved at 89° latitude is around
–120 °C (~153 K). These variations in temperature, as a function of latitude, can be seen in the Lunar
Reconnaissance Orbiter data shown in Figure 2 (Ref. 6).

NASA/CR—2020-220368 3
 TABLE 2.—MOON’S PHYSICAL PROPERTIESa
Declination angle, degree, ψmax ....................................................................................... 1.5
Lunar orbital eccentricity, εl ...................................................................................... 0.0549
Day length, td, h ......................................................................................................... 708.33
Distance from Earth, km
Perigee .................................................................................................................. 363,300
Apogee .................................................................................................................. 405,500
Total days (Earth) to revolve around the Sun ............................................................ 365.25
Orbital eccentricity of Earth/Moon system orbit, ε .................................................... 0.0167
Average orbital radius of Earth/Moon system (about the Sun), rorbm, km ............. 1.496×108
Surface gravity, g, m/s2................................................................................................ 1.623
Lunar to Earth gravity ratio, gr .................................................................................... 0.165
Surface temperature min., K ........................................................................................... 102
Surface temperature max., Tls, K .................................................................................... 385
Surface daytime avg. sink temperature, Tsd, K................................................................ 270
Surface nighttime avg. sink temperature, Tsn, K ............................................................... 84
Lunar radius, rm, km .................................................................................................... 1,737
Lunar surface albedo, als .................................................................................................. 0.3
The mean solar intensity (at Earth’s orbital location), Im, W/m2 .................................. 1,353
aReferences 3 to 5.

Figure 2.—Lunar Reconnaissance Orbiter lunar surface temperature data.

NASA/CR—2020-220368 4
2.1 Surface Characteristics
The cratered highlands and the mare are the two main types of terrain on the lunar surface. The
highlands are the older and significantly cratered parts of the lunar surface and appear brighter from
Earth. The mare are darker regions comprising basaltic lava flows. These areas can be seen in the
topographic map of the lunar surface shown in Figure 3. The blue and purple regions are the lower-lying
mare regions while the red, orange, and yellow areas are the upper highland regions. The lunar regolith
covering the surface of the moon consists of fragments of rocks, minerals, and glass spherules formed
when meteors or other bodies impacted the surface. The regolith thickness varies greatly with location on
the lunar surface. In the mare, the thickness can vary between 3 to 16 m whereas in the highlands it is at
least 10 m thick (Ref. 7). The chemical and/or mineral composition of the regolith is similar to the
underlying bedrock from which it was derived. There is little mixing of material between the highland
and mare regions. The composition of the major materials (>1 percent of the composition) in each of
these regions are given in Table 3.
The regolith is composed of fine grains of the mineral compounds given in Table 2. The grain size of
the regolith varies but consists mostly of particles <1 mm in diameter. This is similar in consistency to silt
or fine sand (Ref. 7). The average regolith density (ρr) is approximately 1,000 kg/m3.

Figure 3.—Lunar Reconnaissance Orbiter lunar topographic map.

Credit: NASA/Lunar and Planetary Institute (LPI). Image processing
by Brian Fessler and Paul Spudis, LPI. (Used with permission.)

TABLE 3.—LUNAR SURFACE

MINERAL COMPOSITIONa
Compound Highland, Mare,
percent percent
SiO2 44.50 41.0
Al2O3 26.00 12.8
FeO 5.77 16.2
CaO 14.90 12.4
MgO 8.05 9.2
TiO2 ------- 7.3
aReferences 7 and 8.

NASA/CR—2020-220368 5
 Figure 4.—Iron concentration on the lunar surface. Credit: NASA/Lunar
and Planetary Institute (LPI). Image processing by Brian Fessler and
Paul Spudis, LPI. (Used with permission.)

Figure 5.—Titanium concentration on the lunar surface. Credit:

NASA/Lunar and Planetary Institute (LPI). Image processing by
Brian Fessler and Paul Spudis, LPI. (Used with permission.)

Key compounds in the production of oxygen from lunar material are the iron oxides and titanium dioxides.
Figure 4 shows the iron content on the lunar surface and Figure 5 shows a map of the titanium content. In both
of these figures, higher concentrations are in red and orange and lower concentrations are in blue.
By looking at Figure 3 to Figure 5, it can be seen that the highest concentrations of iron oxide and
titanium dioxide (ilmenites (FeTiO3)) occur in the low-lying mare regions. Within these mare regions of
high iron and titanium content, ilmenite is present in the loose regolith. About two-thirds of the ilmenite is
in glassy agglutinates and basaltic rock fragments much larger than the common regolith grain size. The
average amount of small grain size ilmenite available in the mare regions is approximately 7 wt% (Rm′, =
0.07) and 5 vol% (Rv′, = 0.05) of the regolith (Ref. 8).
Based on these percentages, the density of ilmenite (ρi) is approximately 1.4 times that of the regolith
or 1,400 kg/m3. The thermal conductivity of the fine-grained ilmenite is estimated to be 0.425 W/(m·K),
which is similar to that of sand (Ref. 9). The specific heat of the ilmenite (cp) is approximated by that of
the mare regolith. The cp as a function of temperature (Ti) is given by Equation (1) (Ref. 9).

NASA/CR—2020-220368 6
 −1,848.5 + 1,047.41log (Ti ) (J/kg·K)
cp = (1)

The regolith properties described above are summarized in Table 4.

2.2 Incident Solar Radiation

With the absence of an appreciable atmosphere, the solar intensity (Is) at the lunar surface is fairly
constant. There is a slight variation throughout the year due to the Earth’s orbital eccentricity about the
Sun. The variation in Earth’s orbital radius (rorb) from the mean orbital radius (rorbm, which has a value of
1.496×108 km) is represented by the eccentricity (ε) of Earth’s orbit that has a value of 0.017 as shown in
Table 2. The distance between the Moon and Earth varies between 360,300 km (at perigee) to 405,500 (at
apogee). The actual solar flux, or Is, in W/m2 for a specific day of the year is determined by Equations (2)
and (3) where γ is the day angle. This variation in Is throughout the year is shown in Figure 6.

 r2 
I s = I m  orbm
2 
(W/m2) (2)
r
 orb 

1 − ε2
rorb = (m) (3)
1 + ε cos ( γ )

TABLE 4.—SUMMARY OF REGOLITH PROPERTIES

Granular ilmenite density, ρi, kg/m3........................................... 1,400.00
Ratio of mass of ilmenite to mass of regolith, Rm ............................. 0.07
Regolith coefficient of friction, µr .................................................... 0.40
Ilmenite particle density, ρip, kg/m3 ........................................... 4,790.00
Regolith and ilmenite surface temperature, Tio, K......................... 384.00

Figure 6.—Variation in solar intensity at a distance of 1 AU.

NASA/CR—2020-220368 7
 The γ used in Equation (3) is defined as 0° on January 4 (perihelion of Earth’s orbit) and increases by
0.98° per day. The solar output varies from approximately 1,400 W/m2 (on January 4) to 1,308 W/m2 (on
July 9). The variation on the Is at the lunar surface due to the orbit of the Moon around the Earth is very
slight. The maximum change in Is resulting from the Moon’s orbit about the Earth is a decrease of about
0.56 percent (~ 6 W/m2) at the lunar surface. As the Moon orbits the Earth (defined as a lunar day), this
small change in intensity will vary. Therefore, it was assumed to be negligible and the Is at Earth’s orbit
was utilized in the subsequent analysis.

3.0 Oxygen Production

There are a number of potential methods for extracting oxygen from the lunar soil. Each process has
significantly different power requirements, both thermally and electrically, as well as very different
infrastructure requirements. Some of the more noted potential processes for removing oxygen from the
lunar surface include

• Hydrogen reduction of ilmenite

• Carbothermal reduction of ilmenite and other oxides
• Hydrogen sulfide reduction of oxides
• Carbochlorination
• Fluorine exchange
• Hydrofluoric acid leach
• Direct electrolytic reduction of oxides
• Caustic solution electrolytic reduction of oxides
• Reduction of metal oxides by lithium or sodium
• Reduction of anorthite by aluminum
• Vapor phase reduction
• Ion separation
• Electrolysis of water ice

Hydrogen reduction of ilmenite is one of the more commonly proposed methods of oxygen production
from lunar regolith and was the method used for this analysis. This is a fairly simple method that can be
accomplished with present day technologies and uses ilmenite, an iron titanium oxide (FeTiO3), as the
oxygen source. Ilmenite is found throughout the lunar surface enabling the process to be utilized at almost
any location. Because the reaction takes place at a relatively low temperature (~1,000 °C), below the
melting point of the soil, the material compatibility and corrosiveness are not as significant an issue when
compared to some of the other methods. The process utilizes hydrogen gas that will need to be supplied
from Earth. However, the gas is mostly recoverable in the oxygen generation process. After the initial
supply, only makeup gas will be needed to account for the gas not recovered or lost through leaking.
Another advantage of this type of production is that it can utilize the surface regolith in areas that have a
high iron and titanium content. This eliminates the need for the power-intensive activities of subsurface
mining or processing rocks. The ability to directly utilize the surface regolith is a significant advantage for
demonstrating an initial oxygen production technology or capability.
The hydrogen reduction of ilmenite is a two-step process represented by the reactions given in
Equations (4) and (5) (Ref. 8). The first step in the process is the reduction of the ilmenite to iron,
titanium dioxide, and water. This reaction takes place at approximately 1,000 °C. The second step is the
electrolysis of the water to form hydrogen and oxygen.

NASA/CR—2020-220368 8
 2FeTiO3 + 2H2 → 2Fe + 2TiO2 + 2H2O (4)

2H2O → 2H2 + O2 (5)

The overall resulting chemical reaction is given by Equation (6).

2FeTiO3 + 2H2 → 2Fe + 2TiO2 + O2 + 2H2 (6)

The resulting oxygen will be stored for later use and the hydrogen will be recycled to perpetuate the
process. The reduction process is performed in a fluidized bed reactor and the electrolysis is performed
with an electrochemical electrolyzer.
A diagram of this process is given in Figure 7. This figure shows a nuclear reactor as a potential heat
source. However, electrical power can drive the heaters within the oxygen reactor to maintain the
temperature of the oxygen reactor. The process requires the ilmenite to remain at an elevated temperature
of 1,273 K for approximately 1 h. An additional benefit of heating the ilmenite is the liberation of
hydrogen. Hydrogen is present in mature lunar regolith because of exposure to the solar wind over long
periods of time. The finer the particle grains, the larger the amount of hydrogen it will have trapped. This
hydrogen content will range from 0.1 to 0.2 kg/m3 of regolith. Since this hydrogen content is dependent
on grain size and not material content, it is reasonable to assume the hydrogen concentration in the
ilmenite is the same throughout the overall regolith. Therefore, as the ilmenite is heated, the hydrogen
bonded to it will be released. It is estimated that at 900 °C, all of the hydrogen trapped with the ilmenite
will be released. This released hydrogen will be utilized in the reduction of the ilmenite and will add to
the overall hydrogen already in the system. It can be considered a source of makeup hydrogen to help
offset any hydrogen losses from the system.

Figure 7.—Oxygen production process and components.

NASA/CR—2020-220368 9
 The sequence for the hydrogen reduction of ilmenite process illustrated in Figure 7 consists of the
following:

• The regolith is scooped and transported to the screen.

• The larger regolith particles (greater than 1 mm in diameter) are separated using a vibrating
screen and removed.
• The screened regolith is transported to a magnetic separator where the ilmenite is separated from
the regolith.
• The ilmenite is transported to the fluidized bed reactor.
• In the reactor, the ilmenite is heated (1,273 K for 1 h) while hydrogen gas is circulated.
• The water vapor output is passed through an electrolyzer where it is separated into hydrogen and
oxygen gas.
• The hydrogen gas is returned to the reactor and the oxygen is stored.

3.1 Oxygen Production System Power Analysis

An analysis was set up to evaluate the sizing of the system shown in Figure 7 for various rates of
oxygen production. To perform the sizing analysis, an energy balance between the heat provided by the
concentrator mirror and the requirements of the various components both thermally and electrically has to
be made. Each of the components’ thermal or electrical energy consumption has to be determined and
scaled with the amount of regolith and/or ilmenite being processed. Since the process involves moving the
material around, the height and material movement rates are needed in order to determine the energy
consumption. These values can be varied to determine what effect they have on the overall system size
and production rate.
The rate of ilmenite entering the reactor sets the rates for all of the other processes. This rate can be
represented by the fraction of the reactor volume that is replenished each hour Vir . This quantity will
have a value between 0 and 1. Establishing the replacement rate of the ilmenite in the reactor also sets the
volume and mass flow requirements for all the other processes in the system. From this, the mass rate at
which the regolith must be acquired from the surface ( M r in kg/h) can be determined. This rate is based
on the total volume of ilmenite in the fluidized bed reactor and is shown in Equation (7). The reactor is
assumed to be cylindrical. Its diameter (dr ) and height (hr,) can be adjusted to increase or decrease the
overall volume of ilmenite processed based on the desired oxygen production rate.

ρ V πd 2 h
M r = i ir r r (kg/h) (7)
4 Rm

The rate at which the regolith is taken from the surface defines the power required to scoop the
regolith (Prs) and transport it to the screen (Prl) with a belt. These quantities are given in Equations (8) and
(9), respectively. The Prs is dependent on the distance the belt scoop travels in the regolith (drs, assumed
to be 1 m) and the coefficient of friction of the regolith (µr). The µr was assumed to be 0.4, which is
similar to that of sand. The power required to lift the regolith depends on the height that the scooped
material is raised plus the Prl. The belt power was scaled using the ratio of the lunar to Earth gravity
constants and an equivalent earth power to operate a 3-m-long, 0.4-m-wide belt at an angle of
approximately 30°. This height is assumed to be the height of the reactor plus the height the reactor is
above the surface (hrs, assumed to be 1 m). The electric motor efficiency (ηem) to drive the scoop and lift

NASA/CR—2020-220368 10
the regolith is assumed to be 90 percent and the fraction of the batch time that is used to scoop the soil
(tsc) is assumed to be 0.01.

µ r d rs M r g
Prs = (W) (8)
3,600tsc ηem

( hr + hrs ) M r g
=Prl + 370 g r (W) (9)
3,600ηem

Once the regolith is picked up from the surface, the first step is to screen out the larger particles; those
greater than a specific size are removed. This screening is necessary for the efficient operation of the
fluidized bed reactor. The maximum particle size (ds) and subsequent screen opening that allows for
efficient fluidization is calculated in Equation (10) (Ref. 8). This equation has to be solved iteratively for
the maximum particle size. It factors in a number of parameters for the operation of the fluidized bed
reactor. These include the porosity of the fluidized bed (εp, estimated to be 0.5) and the hydrogen gas
velocity within the reactor (UH2, estimated to be 0.3 m/s). Details on the hydrogen gas properties and the
ilmenite are also required. These include the hydrogen gas density (ρH2, given in Eq. (11)), hydrogen
viscosity (µH2, estimated to be 2.37×10–5 kg/(m·s)) (Ref. 8), ilmenite particle shape factor (τ, estimated to
be that of sand, 0.83), and the ilmenite particle density (ρip, estimated to be 4,790 kg/m3). The ilmenite
particle density is greater than that of the ilmenite in the soil. This is because the density of the ilmenite in
the soil considers any spacing between the particles whereas the particle density is that of the material that
makes up each particle.

150ρH2 (1 − ε p )U H2 g ρH2 ( ρip − ρH2 ) 2 1.75ρ2H2 U H2 2

= ds − ds (10)
τ2 ε3p µ H2 µ 2H2 τε3p µ 2H 2

The ρH2 is based on the ideal gas law (pressure (PH2) in Pascal, Ti in Kelvin, and the gas constant for
hydrogen (RH2), 4,157 N·m/(kg·K)) with a compressibility factor (ZH2) for the hydrogen gas (Ref. 10).

PH
ρH 2 = 2 (11)
Z H2 RH2 Ti

Z H 2 = 0.99704 + 6.4149 × 10−9 PH 2 (12)

The screening method selected for this analysis is a vibratory screen. A vibratory screening process is
fairly simple. A screen with a specified mesh or opening size is vibrated as the regolith is poured onto it.
The material that is less than the opening size will fall through to the next stage in the process whereas the
material larger than the openings will be moved across the screen surface and discharged from one end of
the screen and returned to the lunar surface. The power required to operate the vibratory screener (Pvs) is
given by Equation (13) (Ref. 8).

0.45M r
Pvs = (13)
507.72d s + 49.91 d s

NASA/CR—2020-220368 11
 Figure 8.—Particle size distribution in the regolith.

The screening process will eliminate a portion of the regolith that was collected. The fraction of the
regolith that will make it through the screen will depend on the desired maximum particle size determined
by Equation (10). The particle size composition fraction (nri) of any size particle (di) within the regolith in
meters is given by Equation (14). The composition fraction represents the fraction of regolith particles at
the specified diameter or smaller within a general regolith sample. This equation is a curve fit
representation of data (obtained from Ref. 8) on the percent concentration of particle size within the
regolith. The particle size distribution along with the corresponding curve fit is plotted in Figure 8.

di − 9.4008 × 10−6
nri = (14)
di + 8.0419 × 10−6

The next step in the process is to separate the ilmenite from the regolith. For a smaller-scale oxygen
production plant, the use of a permanent magnet or electromagnetic separator is preferred. This type of
separator minimizes the power requirements compared to an induced magnetic roll separator or an
electrostatic separator. The electromagnetic separator consists of a drum with alternating magnetic poles
along its surface. Material is dropped over the drum as it rotates. Magnetic material will adhere to the
surface where it is scraped off and captured. Nonmagnetic material will drop out of the bottom of the
separator (Ref. 11). The power consumed by the separator (Pms) in watts is given by Equation (15) (Ref. 8).

Pms = 0.196nri M r (15)

The separated ilmenite has to now be lifted into the fluidized bed reactor. As in Equation (9), this
distance is assumed to be the hr plus the hrs. The power required to lift the ilmenite (Pil) is given by
Equation (16). Where the mass flow of ilmenite ( M i ) in kg/h is given by Equation (17). This mass flow
is based on nri, which is based on ds given by Equation (10) and used in Equation (14) for the particle size.
It is assumed that the separation process is 90 percent efficient (ηs). Therefore 90 percent of the mass flow
given by Equation (17) will be ilmenite. The remainder will be other materials that will not contribute to
the reaction although it will go through the complete heating process.

NASA/CR—2020-220368 12
 ( hr + hrs ) M i g
=Pil + 370 g r (16)
3,600ηem

M i = nri Rm M r (17)

Once the ilmenite enters the fluidized bed reactor, it is heated to 1,273 K and hydrogen gas is
circulated through it. The energy to heat the soil comes from the power source. The concentrated solar
radiation is focused onto a heat receiver. This heat receiver then conducts the heat into the reactor
chamber where it is distributed throughout the chamber to heat the ilmenite. It is assumed that the reactor
is designed to uniformly heat the ilmenite once it is in the reactor. As the ilmenite passes from the top of
the reactor to the bottom, it is slowly heated. The selected rate at which the volume of ilmenite in the
reactor is replaced, given previously as Vir , sets the amount of power needed to heat the soil. The ilmenite
must be brought up to temperature and remain at temperature for a specified period of time (trt, in hours)
in order for the reduction reaction (given by Eq. (6)) to take place. The time available to heat the ilmenite
(th, in hours) is given by Equation (18). From this equation, it should be observed that there is an inherent
limit on the rate at which the ilmenite is replaced and the duration at which it must be at temperature. This
is because the time available to heat the ilmenite must be greater than zero.

 1 
th 
= − trt  where Vir trt < 1 (18)
 Vir 

The thermal power required to heat the ilmenite (Qhi), given by Equation (19), is dependent on this
available heating time, the volume of ilmenite being heated, and the initial ilmenite properties. It is
assumed that the ilmenite is initially at the illuminated surface temperature (Tio) of 384 K.

ρi c p (Ti − Tio ) πd r2 hr (1 − trtVir )

Qhi = (19)
14, 400th

The ilmenite will lose heat to the surroundings (Qls) as it is being heated. To minimize this heat loss,
the outside of the reactor vessel can be insulated with multilayer insulation (MLI). Because of the vacuum
conditions at the lunar surface, radiation is the main heat transfer mechanism in which the reactor vessel
will transfer heat to the surroundings. Therefore, multiple layers of a low emissivity (εi) material (such as
polished nickel, ει = 0.17) that are closely layered will provide very good insulation. The greater the
number of layers (nl) and the smaller the spacing between the layers (dil), the less the heat loss to the
surroundings will be. The heat loss to the surroundings (at a sink temperature Ts = 270 K) from the
insulated reactor vessel can be calculated from Equation (20) where the Stefan-Boltzmann constant (σ)
has a value of 5.67×10–8 W/(m2·K4). If the reactor is constructed of a material different from that of the
insulation, it will have a different emissivity (εrO2, such as chromium εrO2 = 0.14).

π ( d r + nl dil ) hr σ (Ti 4 − Tsd4 )

Qls =
 1 2n  (20)
 + l − nl + 1
 ε rO2 εi 

In addition to the heat required to bring the ilmenite to the desired temperature and the loss of heat to
the surroundings, the hydrogen reduction reaction itself will absorb heat from the system. The amount of

NASA/CR—2020-220368 13
heat consumed (Qr) due to the heat of reaction (Hr) is given by Equation (21). The heat of reaction for the
hydrogen reduction of ilmenite is given as 294 kJ/kg (Ref. 8).

H r ηs M i
Qr = (21)
3,600

The last power-consuming component of the system is the electrolyzer. The electrolyzer’s power
requirement will depend on the amount of water that needs to be electrolyzed. The rate of water
production ( M w ) is based on the ratio of the molecular weights of the reactant and product of interest,
water. The rate of water production in kg/h is given by Equation (22). Subsequently, the rate of oxygen
production ( M o ) can then be determined, which is given by Equation (23).

18
M w = ηs M i (22)
308

16
M o = M w (23)
18
A proton exchange membrane (PEM) electrolyzer was selected to separate the water into oxygen and
hydrogen. The power required by the electrolyzer (Pe) can be calculated based on the theoretical power
needed to break apart water given by the relationship for Gibbs free energy (∆G ) in Equation (24)
(Ref. 12). For liquid water at 340 K (TH2O), the enthalpy change (∆H ) is 285.9 kJ/mol and the entropy
change (∆S) is the change given by Equation (25) where ∆SH2O = 69.94 (J/K)mol, ∆SO2 = 205.29
(J/K)mol, and ∆SH2 = 130.59 (J/K)mol. This results in an entropy change of –0.163 (kJ/K)mol. Therefore,
using the previously calculated ∆H and ∆S, the available ∆G is 230.4 kJ/mol or 3,520 Wh/kg. Using this
value, the mass flow rate of water, and the efficiency of the electrolyzer (ηe), the electrolyzer power can
be determined as given by Equation (26).

∆G =∆H − TH 2O ∆S (24)

∆S = −∆S H 2O + ( 0.5∆SO2 + ∆S H 2 ) (25)

M w
Pe = ∆G (26)
ηe

The electrical power needed to operate the entire system, including the scoop and belt motors, the
magnetic and vibratory separators, move the ilmenite, and operate the electrolyzer, is provided by the
electrical power source. The required power output (PtO2) of the electrical source is given by Equation (27).

PtO2 = Prs + Prl + Pvs + Pms + Pil + Pe (27)

And the total thermal power (QtO2) output needed is given by Equation (28).

QtO2 = Qr + Qls + Qhi (28)

NASA/CR—2020-220368 14
 Figure 9.—Oxygen production rate analysis logic.

Equations (7) to (28) are used to size the oxygen production system and determine its power requirements.
This is an iterative process that is based on selecting the desired oxygen production rate. To start the process of
sizing the system to meet this oxygen production rate, a regolith processing rate is estimated. From this initial
estimate, an oxygen production rate is calculated. Depending on this rate and its variation from the desired rate,
the regolith processing rate is adjusted and the calculation redone. This iteration is continued until the output
rate matches the desired input rate. This process is illustrated in Figure 9.

3.2 Oxygen Production System Power Requirements

Using the analysis described in Section 3.1, the required power, both electrical and thermal, was
determined for the ISRU oxygen production plant. A baseline case was produced that represents the initial
design and operation point for the ISRU system. The baseline assumptions used to generate this case are
given in Table 5. These assumptions represent the best estimates of the performance for the various
components as well as an initial sizing that is within the scope of a demonstration oxygen production system.
Using the assumptions given in Table 5, the total electrical and thermal power required for a range of
oxygen production rates was determined and is shown in Figure 10. A breakdown of the thermal and
electrical power requirements for the baseline oxygen production rate of 1.63 kg/h are given in Figure 11
and Figure 12, respectively. The total electrical power requirement is 9,329 W and the thermal power
requirement is 16,502 W. Figure 11 shows the main thermal requirement for the system is for heating the
ilmenite, as would be expected, which also demonstrates the importance of separating the ilmenite from
the regolith. Since the heat required to bring the material up to the temperature where the reaction can
take place is the largest power-consuming process within the system, minimizing the heating of any
material that does not contribute to the oxygen production reaction is critical to minimizing the power
requirements for the ISRU oxygen production system. Of the electrical power requirements shown in
Figure 12, the electrolyzer uses almost all of the power to break apart water producing oxygen.

NASA/CR—2020-220368 15
 TABLE 5.—BASELINE ASSUMPTIONS FOR
OXYGEN PRODUCTION PROCESS
Desired oxygen production rate, M o , kg/h ....................................................1.63
Total oxygen production over 1 day and night cycle, kg .........................1,154.00
Reactor diameter, dr, m ..................................................................................0.80
Reactor height, hr, m .......................................................................................0.80
Height fluidized bed reactor is above the surface, hrs, m ................................1.00
Distance scoop travels in the regolith, drs, m ..................................................1.00
Regolith and ilmenite conveyor belt length, Lcb, m .........................................3.00
Fraction of batch time to scoop regolith, tsc ....................................................0.01
Electric motor efficiency, ηem .........................................................................0.90
Fluidized bed porosity, εp................................................................................0.50
Hydrogen velocity through the fluidized bed, UH2, m/s ..................................0.30
Ilmenite particle shape factor, τ ......................................................................0.83
Ilmenite hydrogen reduction reaction temperature, Ti, K .........................1,273.00
Ilmenite separation process efficiency, ηs .......................................................0.90
Time ilmenite is at the reaction temperature, trt, h ..........................................1.00
Number of layers of the multilayer insulation (MLI), nl ...............................50.00
Spacing between insulation layers, dil, mm .....................................................0.10
Reactor emissivity, εr ......................................................................................0.17
Insulation layer emissivity, εi ..........................................................................0.17
Electrolyzer efficiency, ηe ..............................................................................0.72

Figure 10.—Required thermal and electrical power over a

range of oxygen production rates.

NASA/CR—2020-220368 16
 Figure 11.—Baseline oxygen production process thermal power requirements breakdown.

Figure 12.—Baseline oxygen production process electrical power requirements breakdown.

The thermal power can also be supplied by electrical heaters powered by an all-electric power system.
For this case, a comparison of all power required by the system, both electrical and thermal, is shown in
Figure 13. This comparison shows that heating the ilmenite is the largest power-consuming element
comprising over 50 percent of the total system power. This power level is somewhat location and time
dependent. There is a large variation in surface temperature depending on the latitude and time of year as
shown in Figure 2. Operating at a time with higher surface temperatures will reduce the amount of heat
needed to bring the ilmenite up to the desired temperature, whereas operating near the poles will require
additional heat due to the cold surface temperatures even while sunlit. For example, operating near the
poles where the surface temperature is 100 K will increase the power required to heat the ilmenite by
32 percent (from 13.4 to 17.7 kW) or a 16.5-percent increase in the total power required by the ISRU
system. This approximately represents the maximum effect that site selection and time of operation can
have on the system power requirement.

NASA/CR—2020-220368 17
 Figure 13.—Baseline oxygen production process total electrical power requirements breakdown.

TABLE 6.—ELECTROLYZER COMPONENT AND

SCALING FACTOR BREAKDOWN (kg/kW)
Electrolyzer stack, Ses ............................................. 2.00
Water tank and heater, Swt ...................................... 0.10
Filters, Sf ................................................................ 0.12
Propellant lines and fittings, Spl.............................. 0.52
Controller unit, Scu ................................................. 0.16
Wiring, Sw .............................................................. 0.30
Heat exchanger, She ................................................ 1.00
Water pump, Swp..................................................... 0.27
Check valves, Schv .................................................. 0.08
Flow regulators, Sfr................................................. 0.62
Control valves, Scv .................................................. 0.16
Pressure and temperature sensors, Ss ...................... 0.07
Flow sensors, Sfs ..................................................... 0.06

3.3 Oxygen Production System Mass Analysis

To accurately model the ISRU oxygen production system and the effect the power system
requirements have on it, its mass must also be considered. To estimate the ISRU system mass, scaling
equations were used for each of the components of the system. These equations allowed the system mass
to be scaled based on their power consumption and regolith or ilmenite mass flow processed.
To size the mass of the electrolyzer (Me), it was broken down into a number of components and scaled
linearly based on its operational power level. The mass of the electrolyzer is given by Equation (29) and the
components along with their scaling factors are listed in Table 6. The scaling values were based on
representative commercially available components over a range of operating parameters as well as projected
performance values for certain items. Linear curve fits for this data were used to generate the scaling factors.
A heat exchanger is used to precool the water vapor to a liquid and filters are used to remove any
particulate matter from the liquid water prior to it entering the electrolyzer.

NASA/CR—2020-220368 18
 M=
e Pe ( Ses + S wt + S f + S pl + Scu + S w + S he + S wp + Schv + S fr + Scv + S s + S fs ) (29)

The electrolyzer system will also require a radiator and its associated equipment to reject its waste
heat during operation. The radiator mass (Mr) is calculated in a similar fashion as that described for the
battery system in Section 5.1. As described in detail in Section 5.1, the required radiator area (Ar) is given
by Equation (77) and its subsequent mass is given by Equation (79). The thermal power that must be
rejected by the radiator (Pte) is given by Equation (30). This power is used as the radiator power to
calculate its size. If the electrolyzer is operating only during the daytime, then louvers will be needed to
reduce the heat loss from the system during the night.

 1 
=Pte 3,520 M w  − 1 (30)
 ηe 

Also described in detail in Section 5.1, the mass of the remaining components of the thermal control
system (MLI, cold plates, and heat pipes) are given by Equations (80) to (82), respectively, and the total
thermal control system mass is given by Equation (81). The specific volume of the electrolyzer (Sve) is given
by Equation (88) and the corresponding area of the electrolyzer enclosure is given by Equation (89).
To minimize the mass of the reactor, hydrogen and oxygen storage tanks were spherical in shape. The
total reactor tank mass (Mrt) is based on the mass of the reactor tank (Mt) itself plus the mass of the
internal components of the tank as given by Equation (31). These internal components would include an
auger mechanism to move the ilmenite while in the tank, lines to channel and distribute the hydrogen gas
within the chamber, and lines to collect and channel the water vapor out of the tank as it is produced. For
this analysis, these components were not sized directly. Their mass was estimated by using a scaling
factor based (Srti) on the reactor tank mass.

M=
rt M t + M t S rti (31)

The tank mass for each of the three tanks used in the system (reactor and hydrogen and oxygen
storage) is determined by calculating the required wall thickness (tt) to withstand the tank’s internal
pressure (Pt) while operating. The tank’s wall thickness is given by Equation (32) which is based on the
yield strength (σy) of the material used to construct the tank and the factor of safety (fs) incorporated into
the tank design.

Pt dt f s
tt = (32)
2σ y

The wall thickness can now be used to calculate the volume of material (Vtm) used to construct the
tank as given by Equation (33).
3
 d d 3
4π  t + tt  −  t  
 2   2  (33)
Vtm =
3

With this volume, the tank mass can then be calculated based on the tank material density (ρtm) as
given by Equation (34).

NASA/CR—2020-220368 19
 M=
t Vtm ρtm (34)

The tank mass calculation, given by Equation (34), can be used to calculate the mass for the reactor
tank as well as the hydrogen and oxygen storage tanks. The variables used to calculate each of the tank
masses are summarized in Table 7.
The mass of the vibrating screen, which is used to separate the particles of the regolith above a
specific size, is based on the size of the particle that is passed through the screen and the rate at which the
material is screened. The screening rate ( Rvs ) as a function of the maximum particle size is shown in
Figure 14 and modeled by Equation (35), which in turn provides the required vibrating screen area (Avs)
given by Equation (36).

TABLE 7.—OXYGEN PRODUCTION TANK SIZING VARIABLES

Variable Symbol Tank
Reactor Hydrogen storagea Oxygen storageb
Pressure, psi (MPa) Pt 147 (1) 1,350 (9.3) 1,350 (9.3)
Baseline tank diameter, m dt 0.8 0.8 1.27
Factor of safety fs 2.5 2.5 2.5
Tank material --- Titanium alloy Carbon composite Carbon composite
Material yield strength, psi (MPa) σy 120,000 (825) 275,000 (1,900) 275,000 (1,900)
aDiameter assumed to be similar to reactor tank.
bDiameter calculated based on the mass of the oxygen produced during 1 lunar day and night cycle.

Figure 14.—Vibrating screen material flow rate versus

particle size.

NASA/CR—2020-220368 20
 −30.76d s2 + 2,653.31d s + 7,836.44
Rvs = (35)

M r
Avs = (36)
Rvs

Using the required vibrating screen area, the mass of the vibrating screen (Mvs) can be estimated. This is
based on the relationship between the vibrating screen area and mass given in Figure 15 and Equation (37).

M vs =36.35 Avs2 + 193.63 Avs + 57.68 (37)

The mass of the magnetic separator (Mms), which is used to separate the ilmenite from the regolith
after it is screened, is given by the relationship between the required separation rate and separator mass
given in Figure 16 and Equation (38).

M ms = 0.0497 Avs2 + 2.279 Avs + 97.53 (38)

The last items of the oxygen production system are the conveyor belts for moving the regolith to the
vibrating screen and the ilmenite to the reactor. It is assumed that both belt systems are the same size and,
therefore, mass. The conveyor belt mass (Mcb) is based on an estimate of the belts used to move rocks and
soil on Earth and scaling it for operation within the lower lunar gravitational environment. This
relationship is given by Equation (39).

M cb = 51.33 g r Lcb (39)

Figure 15.—Vibrating screen mass based on screen

area. Figure 16.—Magnetic separator mass versus the
separation rate.

NASA/CR—2020-220368 21
 The total oxygen production system mass as a function of the production rate is shown in Figure 17.
This figure is based on the mass scaling equations given previously with a storage capacity for 1 day and
night cycle’s oxygen production. The mass breakdown for the baseline rate of 1.63 kg/h is shown in
Figure 18. This figure shows that the largest mass component is the oxygen storage tank. The tank is
sized to store 1 lunar day and night period of the oxygen produced by the system. At the baseline
production rate, this is a total of 1,155 kg of oxygen.

Figure 17.—Mass of the oxygen production system as a

function of the production rate.

Figure 18.—Baseline oxygen production system mass breakdown.

NASA/CR—2020-220368 22
 Figure 19.—Cross section of a small six-person base camp.

4.0 Habitat and Base Camp

A small base camp would consist of a habitat and support equipment needed to support the personnel
and equipment used at the camp, as illustrated in Figure 19. The power requirements for the base camp
can be categorized into three main areas:

1. Life support equipment (heating, cooling, air handling, and CO2 removal)
2. Communications (equipment and displays for communications to and from Earth and to and from
personnel on the surface)
3. Operational equipment (computers, science equipment within the habitat, experiments, and lighting)

In sizing the power requirements for the base camp, it was assumed that the base camp would be
capable of sustaining six people for day and night operation on the lunar surface. As with the ISRU
system, the operational latitude was 30° N.

4.1 Environmental Control and Life Support

The main power requirement (Pls) for the habitat will be for the environmental control and life
support systems (ECLSSs). This is broken down into a number of different power-consuming items that
include air handling and conditioning (Pahc), biomass (Pbm), food storage and processing (Pfd), thermal
cooling components (Pth), waste processing (Pwst), water processing (Pwp), and support for extravehicular
activities (Peva). The total ECLSS power is given by Equation (40). It should be noted that the thermal
system power identified under the habitat is for operation of the cooling system pumps and other items.
Heaters for heating the habitat are not included and, as will be shown at the end of this section, not
necessary since waste heat for ECLSS operation can be utilized to make up some or all of the heater
power requirements.

Pls = Pahc + Pbm + Pfd + Pth + Pwst + Pwp + Peva (40)

Estimates for the day and night power requirements for the ECLSS of a six-person independent
habitat on the lunar surface are given in Table 8.

NASA/CR—2020-220368 23
 TABLE 8.—SIX-PERSON HABITAT ENVIRONMENTAL
CONTROL AND LIFE SUPPORT SYSTEM (ECLSS)
POWER REQUIREMENTS (kW)a
Air (CO2 removal, oxygen generation), Pahc .....................5.85
Biomass, Pbm ..................................................................... 6.10
Food, Pfd............................................................................ 4.27
Thermal, Pth ...................................................................... 1.03
Waste, Pwst ........................................................................ 0.01
Water processing, Pwp ....................................................... 1.29
Extravehicular activity support, Peva .................................2.50
Total, Pls.......................................................................... 21.05
aReference 13.

A thermal analysis of the heat loss that would occur during the nighttime was performed to determine
if heater power was required during nighttime operation or if the waste heat from the operation of the
equipment was sufficient to maintain the internal operating temperature of the habitat. It was assumed that
the habitat was insulated with MLI. The heat loss paths from the habitat that were considered in the
analysis included

• Through the MLI

• From passthroughs and seams in the MLI
• Out of the radiator connections
• Through the support legs of the habitat contacting the surface
• Through the airlock structure

The amount of heat lost through the insulation is dependent on the habitat geometry, environmental
temperatures (desired internal habitat temperature Thi and the nighttime sink temperature Tsn), their type,
and nl. For this analysis, the habitat was assumed to be a hemisphere. The heat loss from the habitat
through the insulation is given by Equation (41). As with the reactor tank, which was also insulated with
MLI, the heat loss is based on the surface area of the habit (Ah) and the emissivity of both habitat wall
surface (εhw) and MLI layers (εi).

Ah σ (Thi4 − Tsn4 )
Qhi =
 1   2nl  (41)
 ε  +  ε  − ( nl + 1)
 hw   i 

The surface area for the hemisphere is dependent on its diameter (dh) as given by Equation (42).

πd h2
Ah = (42)
2
The MLI is very good at resisting heat flow. However, the majority of heat leak through the insulation
occurs from passthroughs and seams (Qps) in the insulation covering. This heat leak is approximated by
Equation (43), which is based on the mean insulation temperature (Tm) given by Equation (44) and
passthrough constants 1 and 2 ( fn and fp), given by Equations (45) and (46).

 0.000136 
=Q ps 0.664  + 0.000121Tm2  f p f n Ah σ (Thi4 − Tsn4 ) (43)
 4σTm 2


NASA/CR—2020-220368 24
 1
 (T 2 + Tsn2 ) (Thi + Tsn )  3
Tm =  hi  (44)
 4 

The fp is based on the presence of passthrough area (Apt) of items such as wires or tubes that pass
through the insulation. This area is given in percent value, for example, if the estimated passthrough area
is ½ percent, 0.5 is used as the percent passthrough area.

f p 0.73 + 0.27 Apt

= (45)

=f n 4.547 − 0.501ni (46)

The values used for the variables in determining the heat loss through the passthroughs and seams are
given in Table 9.
During the night, the coolant loop will be turned off to minimize the heat loss from the radiator. Since
there is no longer any heat flow to the radiator, it will drop in temperature to match that of the
surroundings over the night period. Therefore, the heat leak from the habitat to the radiator (Qra) will be
dependent on the conduction down the coolant tubes that connect the radiator to the habitat. This is
similar to the heat loss that occurs through the support legs (Qsl) and the airlock structure (Qas). The heat
leak (Qcp) through conduction from these sources is given by Equation (47), which is dependent on the
number of conductive paths (ncp), the thermal conductivity of the material (k), the cross-sectional area of
the material normal to the direction of the heat flow (Acp), and the length of the conductive path (Lcp).

ncp kAcp (Thi − Tsn )

Qcp = (47)
Lcp

The cross-sectional area for the conductive paths is given by Equation (48). It was assumed that all of
the paths considered could be represented by a hollow cylinder shape with a specified inner diameter (dicp)
and wall thickness (tcp) where the cross-sectional area of that shape is normal to the flow of heat from the
interior of the habitat to the surroundings.

2 2
 dicp   dicp  
Acp =
π  + tcp  −    (48)
 2   2  

The variables used to determine the heat leak for each of the identified conductive paths are
summarized in Table 10.

TABLE 9.—HABITAT PASSTHROUGH AND

SEAMS HEAT LEAK VARIABLES
Habitat wall emissivity, εhw ......................... 0.07
Insulation emissivity, εi ............................... 0.07
Number of layers of insulation, nl ............. 25.00
Habitat diameter, m, dh ............................... 8.00
Percent of passthrough area, Apt .................. 1.00

NASA/CR—2020-220368 25
 TABLE 10.—HABITAT CONDUCTIVE PATH HEAT LEAK VARIABLES
Variable Symbol Radiator coolant lines Support legs Airlock structure
Number of conductive paths ncp 16 10 2
Material --- Stainless steel Titanium alloy Stainless steel
Ti-6Al-4V
Thermal conductivity, W/(m·K) k 19.00 6.70 19.00
Inner diameter, m dicp 0.02 0.46 1.48
Thickness, cm tcp 0.10 2.00 10.00
Length, m Lcp 0.10 1.00 0.25
Interior temperature, K Thi 295.00 295.00 295.00
Surrounding sink temperature, K Tsn 84.00 84.00 84.00

TABLE 11.—HABITAT NIGHTTIME

HEAT LOSS SUMMARY (W)
Insulation, Qi ................................. 245
Passthroughs and seams, Qps ...... 1,883
Radiator, Qra.................................... 42
Support legs, Qsl ............................ 426
Airlock structure, Qas ................. 1,500
Total, Qh ..................................... 4,096

The total heat loss from the habitat to the surroundings during nighttime operation (Qh) is given by
Equation (49) and summarized in Table 11.

Qh =Qi + Q ps + Qra + Qsl + Qas (49)

Since the power required to make up the heat loss (Table 11) is less than that of the ECLSS (Eq.
(50)), the additional heater power (Pht) needed to maintain the habitat at its operating temperature is set to
zero. The waste heat from the other systems can be used to maintain the habitat temperature during the
nighttime period.

Qh < ( Pls − Pth ) ∴ Pht =

0 (50)

4.2 Communications
To support the surface operations, a communications station, as shown in Figure 20, will be needed
by the base camp to provide a data link between the base camp on the lunar surface, orbiting satellites,
and Earth. Examples of the data rate communications requirements are shown in Table 12.
To provide the types of communications links shown in Table 12, the base camp communications
system will be required to both transmit and relay data to points on the surface or in orbit. This type of
system would be very similar to a cellular phone transmission tower system. As with the habitat itself, the
communications system will need to operate continuously throughout the day and night.

NASA/CR—2020-220368 26
 Figure 20.—Lunar surface communications station.

TABLE 12.—COMMUNICATIONS REQUIREMENTS

FOR VARIOUS LUNAR APPLICATIONSa
Application Data rate, Transmission distance,
Mbps km
Lunar surface-to-spacecraft relay 1,000 6,500
Lunar surface low-rate communications 10 2,700
Surface-to-lunar science orbiter 100 2,700
Lunar surface human outpost 1,000 2,700
aReference 14.

For surface-to-surface transmissions, the required power level for the transmitter (Ptm) and the signal
strength at the receiver must be determined (Pss). From the receiver power level, the maximum
transmission data rate can be determined. For transmission over the lunar surface, the power at the
receiver is given by Equation (51) (Ref. 15), which is dependent on the transmission centerline frequency
(fcl), the transmitter height above the surface (ht), the receiver height above the surface (hre), and the
distance between the transmitter and receiver (dtr). The value for the speed of light (c) is 2.998×108 m/s.

2πf cl ht hre
Ptm c sin
cdtr (51)
Pss =
2πdtr f cl

NASA/CR—2020-220368 27
 The maximum transmission distance (dtrmax) will be limited by the curvature of the lunar surface. This
distance is given by Equation (52) where the radius of the Moon (rm) is 1,737.5 km. Due to interaction
with the surface, the signal can, however, travel beyond the line of sight distance. This increase in
transmission can be represented by an adjustment factor (K = 1.33) (Ref. 4).

 r 
dtrmax = Krm cos −1  m  (52)
 rm + ht 

The receiver power for a transmission from the surface to an orbiting satellite (Pro) is based on
the free-space loss due to dispersion of the signal as it travels from the transmitter to the receiver.
Equation (53) calculates the surface-to-satellite receiver power.

Ptm c
Pro = (53)
4πdtr f cl

The maximum achievable data rate (Rmax) can be calculated based on the receiver power as shown in
Equation (54). Expressed as thermal noise, the data rate is based on the bandwidth of the transmission
(Bw) and the noise power level (Pn) illustrated in Equation (55) (Ref. 16). The thermal noise is a function
of temperature of the transmission and receiving system and the bandwidth of the transmission. For this
analysis, the daytime sink temperature is used as a worst-case thermal noise temperature. The Boltzmann
constant (kb) has a value of 1.3803×10–23 J/K.

 P 
Rmax Bw log 2 1 + ro 
= (54)
 Pn 

Pn = kbTsd Bw (55)

The baseline assumptions used to determine the data rate capability and power requirements for the
communications system are given in Table 13. From these assumptions, the required transmitter power
and corresponding maximum data transfer rate were calculated for the lunar surface-to-lunar orbit, lunar
surface-to-lunar surface, and lunar surface-to-Earth communications and are shown in Figure 21 to
Figure 23, respectively. For the surface-to-surface communications, the maximum transmission distance,
as calculated by Equation (51), was 11.1 km.
Based on the data rates shown in Table 12 and the communications power (Pcom) required to achieve
these data rates for both communications from the lunar surface (to orbit or Earth), a power level of 1 kW
was assumed for the base camp communications system.

TABLE 13.—COMMUNICATIONS SYSTEM BASELINE

Transmitter and receiver tower height, ht, m ........................10
Center operating frequency, fcl, GHz....................................30
Effective noise temperature, Tsd, K ....................................270

NASA/CR—2020-220368 28
 Figure 21.—Transmission power and data rate for lunar surface-
to-lunar orbit (6,500 km) transmission bandwidths.

Figure 22.—Transmission power and data rate for lunar surface-

to-lunar surface (11.1 km) transmission bandwidths.

NASA/CR—2020-220368 29
 Figure 23.—Transmission power and data rate for lunar surface-
to-Earth (384,400 km) transmission bandwidths.

4.3 Operational Equipment

Operational equipment within the habitat includes all of the power-consuming items that are utilized
by the personnel as well as any experimental equipment and general habitat operations and maintenance
equipment. The types of items that would consist of the operational equipment include

• Lights
• Computers
• Networking equipment
• Cameras
• Displays
• Experiments
• Sensors
• Food preparation
• Maintenance equipment (tools, vacuum, etc.)

Data taken at the Desert Research and Technology Studies (Desert RATS) campaign was used to
estimate the operational power for these components. Under this program, an operational habitat
demonstration unit (HDU), as shown in Figure 24, was constructed and operated for a number of weeks at the
Black Point Lava Flow outside of Flagstaff, Arizona. The Desert RATS program was conducted annually
with the goal to provide analog testing of different mission elements operating together in an environment
similar to that of an actual lunar mission to evaluate the interactions between the different elements.
The power distribution system within the Deep Space Habitat (DSH) consisted of six power
distribution units (PDUs). Each PDU had 12 controllable outlets. The power flow from each outlet was
monitored to track the power consumption of individual items or systems as well as the total power
consumed by the habitat during operation. The output power throughout the day for each day of operation
of the habitat during one of the testing periods is shown in Figure 25. This power consumption represents
all of the equipment utilized within the habitat and is a good representation of the operational equipment
power usage for a small habitat.

NASA/CR—2020-220368 30
 Figure 24.—Deep Space Habitat (DSH) testing at Desert Research and
Technology Studies (Desert RATS) 2011. Extravehicular activity is EVA.

Figure 25.—Daily Deep Space Habitat (DSH) power consumption.

NASA/CR—2020-220368 31
 Figure 26.—Base camp power breakdown. Environmental control and life support
system is ECLSS.

The data shown in Figure 26 can be used to estimate the operational equipment power (Poe)
requirements for the habitat. The data shows a power consumption range of 2.5 to 7.0 kW over the total
period of operation of the habitat. Based on this, a conservative power consumption estimate of 6 kW was
selected to represent the ancillary operational equipment power level.

4.4 Base Camp Total Power Requirement

Using the power estimates for the different components that make up the base camp, a total power
requirement (Pbc) can be determined as given by Equation (56). This power requirement is continuous
through both the day and nighttime periods. Since the habitat and its support equipment for the base camp
have to sustain the six-person crew, there is no decrease in power during the nighttime period. The power
requirement summary for the base camp is shown in Figure 26.

Pbc = Pls + Pht + Pcom + Poe (56)

It should be noted that there was no heater power required for the habitat. The heat leak from the
habitat during the nighttime was estimated to be 4,096 W. However, the waste heat generated by the other
systems within the habitat (ECLSS and operations) is much greater than this level. Therefore, it is
assumed that the waste heat can be utilized to maintain the internal habitat temperature during the night at
the desired 295 K without the need for heaters.

5.0 Power System Sizing

To provide power for the ISRU oxygen production system and the base camp, both a PV array-based
power system and Kilopower reactor-based system were considered. The power systems were sized to
meet the electrical and thermal power requirements for both applications. While the power systems were
both sized to operate within the vicinity of the base camp and ISRU system, the Kilopower system was
also sized to operate a distance away to minimize the shielding requirements.

NASA/CR—2020-220368 32
 Figure 27.—Photovoltaic (PV) array-based power system layout. Regenerative fuel cell (RFC).

5.1 Photovoltaic Array Power System

The PV power system consists of a fixed PV array and energy storage system, either a battery or
RFC, to meet the needs of the ISRU system and base camp as illustrated in Figure 27. An RFC system
consists of an electrolyzer as described previously in Section 3.0 and a fuel cell, which is used to provide
the nighttime power output. The fuel cell is an electrochemical device that produces electricity through
the electrochemical recombination of hydrogen and oxygen. It does not “burn” the hydrogen as a
conventional engine would. Instead, it combines it electrochemically to produce electricity and water as
given by Equation (57). The water is subsequently stored and then broken apart into hydrogen and oxygen
gas by the electrolyzer, reversing the direction of the reaction given by Equation (57), during the day
when solar power is available.

2H2(g)+O2 → 2H2O (57)

PV power systems are the most common type of space-based power system currently used. This type
of system’s advantages and disadvantages for the ISRU and base camp power include
Advantages:

• High technology readiness level (TRL) mainly for the PV and battery system; RFC system is
lower.
• Power production is applicable to other mission aspects and will likely be utilized on other
vehicles or installations. This provides a synergy in the power source design and the commonalty
of the power system application could have additional benefits, such as cost, repair parts, and
multiple system integration.

Disadvantages:

• Depending on location and power requirements, it may be limited to daytime operation only due
to the excessive energy storage requirements of operating during the 14-Earth-day lunar night.
• Large size to achieve high power levels.
• The electrical conversion to heat reduces the overall efficiency of systems such as the oxygen
production plant that requires significant amounts of high-quality heat.

NASA/CR—2020-220368 33
 The PV system sizing is heavily dependent on its operational environment. Items such as the time of
year, latitude, and array orientation significantly influence its output power and sizing. For the PV array
sizing, it was assumed that the array was south facing and fixed at a given inclination angle. The array
was sized to provide enough power to meet both the day and nighttime power requirements specified for
each of the cases examined. This was accomplished through an energy balance between the energy
collected by the array during the day and the energy stored in the battery or RFC system for use at night.
The array did not track the Sun, therefore the output power of the array varied throughout the day due to
the changing Sun angle relative to the array surface. The fixed array was chosen since it utilized a simple
lightweight structure and is less complicated to install and operate than a tracking array. It was also
assumed that no applicable dust coverage was allowed to accumulate on the array surface. The power
system assumptions used to size the array are given in Table 14.
The eccentricity of the Earth and Moon system orbit about the Sun causes the Is to vary throughout
the year, as shown in Figure 6. The vernal equinox was selected as the baseline operational date with an Is
of 1,359 W/m2.
It is assumed that the array is at a fixed inclination angle and facing south (for Northern Hemisphere
operation). The input to the array will consist of the direct beam component normal to the array surface as
illustrated in Figure 28. Since there is no atmosphere, there is no diffuse component of the solar radiation.

TABLE 14.—VARIABLES USED TO SIZE SOLAR ARRAY POWER SYSTEM

Latitude..................................... 30° N was selected as the operational latitude for all cases
Date .................................................................. March 21 (vernal equinox), day number 80
Solar cell efficiency, ηsc, percent ...................................................................................... 28
Battery discharge efficiency, percent ........................................... Up to 99 (rate dependent)
Battery charge efficiency, ηbc, percent ............................................................................. 99
Fuel cell efficiency, ηfc, percent ....................................................................................... 65
Electrolyzer efficiency, ηe, percent................................................................................... 85
Array inclination angle and orientation, β ................................................. 30°, south facing
Array solar cell fill factor, fsc ......................................................................................... 0.89

Figure 28.—Solar array orientation on lunar surface.

NASA/CR—2020-220368 34
 The array output power (Pa) can be calculated from Equation (58), which is based on the solar cell
efficiency (ηsc), the area of the solar array (Aa), and the fill factor (fsc) or ratio of the solar cell area to the
array area.

Pa= (1 − χ d ) f sc ηsc Aa I s sin ( α + θ ) (58)

From a fixed point on the lunar surface, the rotation of the Moon causes the Sun to move in an arc
through the sky above that point. That arc describes the elevation angle of the Sun from that fixed point
on the surface. The solar elevation angle (α), given by Equation (59), varies throughout the day from
dawn to dusk at a given latitude (φ).

π   2πti 
α= − cos −1 sin(φ)sin(ψ ) − cos(φ) cos(ψ ) cos   (59)
2   td 

The lunar declination angle (ψ) throughout the year is given by Equation (60). The ψ is much less
than that of Earth. Therefore, the seasonal variation in solar elevation angle is much less on the Moon
than on Earth. Since the Earth and Moon system orbits the Sun with the same period, the day number (dn)
to indicate the position in orbit about the Sun and the total number of days (dnt) needed to make one
revolution about the Sun are based on Earth days.

 2πd n 
ψ = ψ max sin   (60)
 d nt 

The angle of the incident solar flux onto the array surface is a combination of the hour angle (θ) and
the α. The θ is given by Equation (61) which is based on the inclination angle of the array (β) and the
ratio of the instantaneous lunar day time (ti in hours) to the total hours in the lunar day (td from Table 2).

  2πti 
=θ tan −1  tan(β)cos  − π  (61)
  td 

The maximum α for each day throughout the year is plotted in Figure 29 for a number of latitudes.
From this figure, it can be seen how small the variation is throughout the year.
Dust buildup on the array, represented by the percent of dust coverage (χd), would reduce the output
of the solar array over time. However, since this is a manned base camp, it was assumed that the array
would be periodically cleaned and therefore there would be no significant dust buildup during the
operation. The percent of dust coverage was assumed to be zero (χd = 0).
The ηsc of 28 percent was selected as representative of near-term solar array efficiency capabilities for
large-scale space-based arrays. Current research cells can achieve efficiencies greater than this for single
samples under laboratory conditions as shown in Figure 30. However, for large-scale production and use
in a large array system, the overall efficiency will be less than the research sample cells. The efficiency
and areal density (ρa) of different array and solar cell types are listed in Table 15. It should be noted that
the efficiencies listed are the overall array efficiency, which accounts for the fill factor that is used to
adjust from the cell efficiency to the array efficiency.

NASA/CR—2020-220368 35
 Figure 29.—Maximum daily solar elevation angle throughout
year at various latitudes on lunar surface.

TABLE 15.—EFFICIENCY AND AERIAL DENSITY OF DIFFERENT

ARRAY AND PHOTOVOLTAIC (PV) CELL COMBINATIONSa
Array typeb Cell type Array efficiency, Areal density,
percent kg/m2
UltraFlex Thin film 11.5 0.80
UltraFlex Multijunction 25.0 1.59
RAPDAR Thin film 8.5 0.33
RAPDAR Thin film 11.5 0.42
RAPDAR Multijunction 25.0 1.79
Ridged panel Multijunction 25.0 3.12
aReference 18.
bUltraFlex (AEC-Able Engineering Company, Inc.). Rollout and passively
deployed array (RAPDAR).

NASA/CR—2020-220368 36
Figure 30.—Research photovoltaic (PV) cell efficiencies over time from National Renewable Energy Laboratory
(NREL) part of the Department of Energy (DOE) (Ref. 17).

The Aa needed to provide the necessary output to meet the day and night power requirements is
determined through an energy balance between the array output and the day and nighttime power
requirements. An example of this type of energy balance is shown in Figure 31. The energy produced by
the solar array (Ea) is equal to the area under the array output curve, as given by Equation (62). Similarly,
the energy required by the loads during the day (Ed) is the area under the load power curve for the
daytime period, as given by Equation (63) and is based on the daytime operating time for the loads and
their power level (Pdl). For this analysis, it is assumed that the daytime load is constant at any given time
(ti) over the time period it is operational.
td
Ea = ∫ Pa dti (62)
0

td
Ed = ∫ Pdl dti (63)
0

The nighttime power is provided by the energy storage system. The energy stored in either the
batteries or the RFC system is the area under the nighttime power (En) curve, as given by Equation (64),
and is based on the nighttime power loads (Pnl).
td
En = ∫ Pnl dt (64)
0

NASA/CR—2020-220368 37
 Figure 31.—Baseline production-rate energy balance for operating oxygen production plant
continuously day and night.

Therefore, to be able to recharge the energy storage system (batteries or RFC) from the array, the
array has to produce excess energy, beyond what is used for daytime operation, equivalent to what is
consumed from the batteries during the night multiplied by the charge (ηc) and discharge (ηd) efficiencies.
The energy produced by the solar array used to charge the batteries is given by the area under the array
output curve minus the area under the daytime power curve. This relationship is given by Equation (65).

E=
n ( Ea − Ed ) ηc ηd (65)

This relationship is solved by iterating on the array area until the nighttime energy requirement (left
side of the equation) is equivalent to the excess energy produced during the day, factoring in the charge
and discharge inefficiencies (right side of the equation). This approach, iterating on the array area, is
utilized if the day and nighttime power levels are similar, as in the power requirements for the base camp
or if the oxygen production plant was operated continuously day and night.
There are other operational approaches for situations where a large discrepancy in the day and nighttime
power requirements exists. This discrepancy would occur, for example, if the oxygen production plant was
operated only during the day in order to reduce the need for larger nighttime energy storage. While the
nighttime power requirements would be minimal, just enough to provide “keep-alive” power for the
equipment, the plant would have to operate at a much higher production rate to produce the same output of
oxygen. In this situation, the goal would be to match the total energy required (as if the oxygen production
system was running continuously day and night) with an equivalent amount of energy being produced
during only a portion of the daytime period. This operational situation is shown in Figure 32. To accomplish
this, the array is oversized to provide all of the daytime power to the oxygen production system running at
the higher production rate and its corresponding higher power level than the system designed to operate
continuously day and night. Since the goal is to match the total energy of the ISRU process, the iteration is
based on both the array area and the operational power level. The array area sets the total array output’s
power curve, and the operational power level sets the total usable energy for the process within that power
curve. The majority of the array output during the morning and evening is less than the operating power
level of the oxygen production plant, which prevents it from being utilized. Also, the majority of the power
produced during midday, in excess of the amount needed for the oxygen production plant, is not utilized.
Some of this excess power is used to recharge the batteries for the nighttime keep-alive power, but this
amount is minimal compared to the total excess power available. To solve this situation’s energy balance
discrepancy, a two-step iteration on the required operational power level and the total array area is required.

NASA/CR—2020-220368 38
 Figure 32.—Baseline production-rate energy balance for operating oxygen production plant
during only the daytime.

Based on the required array area determined through the energy balance, the mass of the array can be
calculated. The array blanket mass (Mapv), given by Equation (66), is also based on the areal density of the
array given in Table 15. For this analysis, the UltraFlex multijunction array (AEC-Able Engineering
Company, Inc.) areal density of 1.59 kg/m2 is used.

M apv= Aa ρa (66)

The array structure mass (Mas) was based on a fixed carbon composite structure. The structure had an
estimated specific mass (ρas) of 0.55 kg/m2 based on the array area (Refs. 19 and 20). The array structure
mass is given by Equation (67).

M as= Aa ρas (67)

The wiring mass (Mw) to connect the arrays is based on the number of wire runs between the arrays
and junction box (nw) given by Equation (68). The average spacing between the arrays (Sa) was estimated
to be 3 m with the average length of each of the wire runs (Lw) given by Equation (69).

2
 A 
nw =  a  (68)
 Sa 

Aa
Lw = (69)
2
The specific mass of the wire (Msw) in kg/m is given by Equation (70), which is a curve fit of the data
in Figure 33. The Msw is based on the required wire gauge (Gw). The Gw is in turn based on the current
level (Iw) per wire run. This current is given by Equation (71) for an array bus voltage (Va) and total
current to accommodate the Pa. Depending on how the system is operating, to size the Gw, either the Pdl or
the maximum array output power (Pamax), determined from the array output power curve, is used. If the
power system is operating as shown in Figure 31 where the total output energy of the array is utilized to
operate the loads and recharge the energy storage system, then the wire is sized for maximum array
output. If the system operates as shown in Figure 32, where the maximum power transmission is based on
the daytime load power demand, then the daytime load power is used.

NASA/CR—2020-220368 39
 Figure 33.—Power transmission wire specific mass versus wire gauge.

Figure 34.—Required wire gauge for maximum current level.

− ( 4.671 × 10−5 ) Gw3 + ( 2.842 × 10−3 ) Gw2 − 0.0592Gw + 0.4302

M sw = (70)

Pa
Iw = (71)
Va nw

The corresponding Gw needed to carry the required Iw as calculated by Equation (71) is shown in
Figure 34 and corresponding Equation (72). The Iw versus Gw relationship is for a wire in a vacuum
environment.

NASA/CR—2020-220368 40
 −5.911ln ( I w ) + 27.963
Gw = (72)

Since wire gauges are given in whole numbers, the output of Equation (72) is rounded up to the next
highest whole number based on the Iw. This also provides a small amount of margin in the wire size
calculation and provides a conservative estimate of the required Gw.
Once the Gw is known, the Msw can be determined. The Msw is based on using bare, uninsulated wire.
For operation in vacuum on the lunar surface, the bare wires will need to be held apart with spacers at
increments along their length. A benefit of bare wire is the amount of current that can be carried by the
wire since it will reject heat to the surroundings at higher rates than insulated wires. The wires can be
painted or coated to provide a high emissivity for enhanced heat rejection to the environment. In addition,
there are several other drawbacks to using insulated wire. These include the increased mass of the wire
due to the added insulation, insulation damage due to long-term ultraviolet (UV) exposure on the lunar
surface, and the thermal expansion mismatch between the insulation and the wire material, which could
cause failure of the insulation due to the large temperature swings seen on the lunar surface.
The corresponding total Mw for power transmission wire mass can now be calculated from the Msw,
nw, and Lw as given by Equation (73).

M w = M sw nw Lw (73)

The power wires coming from the array are run to a breaker switch box and then connected to the
remainder of the system. The mass of the breaker box (Mbb) is scaled from commercially available
breaker boxes based on the number of connections the box can handle. This relationship is shown in
Figure 35 and expressed by Equation (74).

=M bb 9.6736ln ( nw ) − 13.897 (74)

Figure 35.—Breaker box and switch mass versus number

of wire connections.

NASA/CR—2020-220368 41
 Figure 36.—Battery charge controller mass as a function of
battery operating power.

For the power system utilizing a battery as the energy storage medium, the mass of the battery charge
controller (Mbcc) can be estimated from data on commercially available battery charge controllers. The
battery charge controller mass is based on the operating power level of the battery. This relationship is
shown in Figure 36 and expressed in Equation (75).

−1.756 × 10−7 Pnl2 + 2.035 × 10−3 Pnl − 0.09288

M bcc = (75)

The energy storage mass (Mes) consists of either the battery mass (Mb) or the RFC system mass (Mrfc).
The Mb is based on the specific energy (in Wh/kg) of the battery type utilized for energy storage. The specific
energy ranges for various battery types are shown in Figure 37. The sodium sulfur battery, which has the
highest specific energy, is a type of thermal battery and is not commonly used for long-duration operation.
Therefore, it is not applicable for use during the lunar nighttime. The lithium-ion battery has space heritage
and is commonly used in aerospace applications, which is why it was chosen for use as an energy storage
medium for the PV array and powering the loads during the lunar night. Based on the lithium-ion battery
specification shown in Figure 37, a battery specific energy (Sb) of 200 Wh/kg was used for the battery. This
value represents the specific energy of a packaged battery system comprising multiple cells. The battery mass
also depends on the maximum acceptable battery depth of discharge (Bdod) during anticipated operation. The
Bdod represents the percentage of energy that can be extracted from the battery relative to its total energy
capacity. The Bdod is directly related to the lifetime charge and discharge cycles the battery can achieve as
shown in Figure 38. Although the nighttime periods that the battery must operate through are very long, the
expected number of cycles for the system are very low. For an estimated 15-yr lifetime, the batteries would
experience only 196 cycles. Therefore, to reduce battery mass, the batteries can use a high Bdod of 80 percent
or 0.8. Using this battery type’s specific energy, the required Mb to meet the nighttime lunar energy
requirements is given by Equation (76). It should be noted that due to the low number of operating cycles and
prolonged periods of operation, the battery lifetime may be affected by other failure mechanisms such as
material degradation. However, estimating these other failure mechanisms is out of the scope of this study.

En
Mb = (76)
Sb Bdod

NASA/CR—2020-220368 42
 Figure 37.—State-of-the-art battery specific energy for various battery types (Refs. 21 and 22).

Figure 38.—Lithium-ion battery depth of discharge effect on cycle life (Ref. 23).

NASA/CR—2020-220368 43
 The last component of the battery system is the thermal control to maintain the battery and charging
electronics within their operating temperature range. The thermal control system consists of the following
components:

• Radiator
• MLI
• Heat pipes
• Cold plates

The required Ar, as given by Equation (77), is determined by the thermal power that needs to be
rejected (Pr). The radiator is sized using given variables (described in Table 16) based on the worst-case
operating conditions.

Pr Fal
Ar = (77)
σ( ε r Tr4 − α rls flsTls4 ) − α rs I s ( cos ( θrs ) + fls als )
Since the worst-case operating condition for the radiator sizing will occur during the daytime, the Pr
used to size the radiator is based on the power and efficiency used to charge the battery. The charging
power is the difference between the Pamax obtained from the array output on the energy balance diagram
and the Pdl. Using the energy balance diagram shown in Figure 31, the heat that has to be rejected by the
radiator is given by Equation (78).

ηbc ( Pamax − Pdl )

Pr = (78)

An estimate of the Mr can be made based on its required area. The radiator structure can be separated
into a number of components with a scaling coefficient for each component to linearly scale the mass
based on the required Ar. These coefficients, listed in Table 17, were derived from satellite and spacecraft
radiator mass data. The total Mr is given by Equation (79).

M r = C p Ar + Cc Ar + Ct Ar + Ch Ar + Ca Ar + Cs Ar + Cat Ar + Clv Ar (79)

The last term in Equation (79) represents the louvers. Louvers are used to insulate the radiator when it
is not in use during nighttime operation. They are required to reduce the heat loss to the environment
during this time. Utilizing louvers on the radiator will increase the required Ar needed, due to the reduced
view factor of the radiator to deep space or other cold surfaces to reject heat. This is represented by the
louver area adjustment factor (Fal) given in Table 16 and is estimated to increase the Ar by 30 percent.

TABLE 16.—RADIATOR SIZING VARIABLES

Emissivity, εr....................................................... 0.84
Temperature, Tr, K ............................................... 300
Lunar surface absorptivity, αrls ............................. 0.3
View factor to lunar surface, fls ............................. 0.5
Solar absorptivity, αrs.......................................... 0.14
Maximum Sun angle to radiator, θrs, deg ............... 75
Louver area adjustment factor, Fal ........................ 1.3

NASA/CR—2020-220368 44
 TABLE 17.—RADIATOR MASS
SCALING COEFFICIENTS (kg/m2)
Panels, Cp .................................................. 3.30
Coating, Cc ................................................ 0.42
Tubing, Ct.................................................. 1.31
Header, Ch ................................................. 0.23
Adhesives, Ca ............................................ 0.29
Stingers, Cs ................................................ 1.50
Attachment, Cat ......................................... 0.75
Louvers, Clv ............................................... 4.50

TABLE 18.—MULTILAYER INSULATION

MASS SCALING COEFFICIENTS (kg/m2)
Spacer, Csp............................................... 0.0063
Reflective layer, Crl .................................0.0550
Outer cover, Coc ......................................0.1100
Inner cover, Cic........................................ 0.0500
Attachment and seals, Cas........................0.1000

The insulation mass to enclose the batteries is based on the surface area of the battery enclosure (Abe).
The specific volume of the selected lithium-ion batteries (Svb) is estimated to be 5.5×10–4 m3/kg (Ref. 24).
Using the battery’s specific volume and required Mb, the Abe, which is also the insulation surface area, can
be determined, as given by Equation (80).
2
Abe = 6 ( M b Svb ) 3 (80)

The corresponding mass of the MLI (Mmli) is based on the area to be insulated and scaling
relationships for the components that make up the MLI as given by Equation (81). The scaling
coefficients for the insulation are given in Table 18.

M mli= {[( Csp + Crl ) nl + Coc + Cic ] Abe } (1 + Cas ) (81)

The cold plates and heat pipes are used to remove the heat from the batteries and transfer it to the
radiators. The number of cold plates needed and their mass (Mcp) can be estimated based on the surface
area of the battery enclosure as given by Equation (82) with the variables used to size the cold plates
given in Table 19.

M cp AbeCcf tc ρc
= (82)

The last component of the thermal system is the heat pipes used to move the heat from the cold plates
to the radiators. It is assumed that the radiators are located just outside of the battery enclosure. Therefore,
the average heat pipe length is estimated to be half the distance from the center of the enclosure to the
outer wall. The mass of the heat pipes (Mhp) is given by Equation (83), and the heat pipe (Chp) and cold
plate (Ccf) coefficients are given in Table 19.

Ccf Chp Abe

1.5
M hp = (83)
1.225wc Lc

NASA/CR—2020-220368 45
 TABLE 19.—COLD PLATE COMPONENT
MASS SCALING COEFFICIENTS
Cold plate coverage coefficient, Ccf .................. 0.25
Cold plate thickness, tc, m ............................... 0.005
Cold plate width, wc, m ....................................... 0.1
Cold plate length, Lc, m....................................... 0.1
Cold plate material ................................. Aluminum
Cold plate density, ρc, kg/m3 ........................... 2,700
Heat pipe mass coefficient, Chp, kg/m ............... 0.15

The relations for the individual thermal control system component masses can then be summed to
provide the total thermal control system mass (Mtc) needed to maintain the battery temperature at its
desired level as given in Equation (84).

M tc =M r + M mli + M cp + M hp (84)

The array power system using the battery as the energy storage medium consists of the PV array
blanket and structure, battery charge controller, wire, junction box, and battery. The mass of these
components, as previously outlined, are summed to provide the total array and battery power system mass
(Mab) as given by Equation (85).

M ab = M apv + M as + M bcc + M w + M bb + M b + M tc (85)

The mass of the RFC system consists of the reactants masses with significantly smaller contributions
from the mass of the fuel cell (Mfc) and the Me. The Me is calculated in the same manner as that given in
Section 3.3 for the electrolyzer used with the oxygen production system. For use with the PV array
system, its sizing is represented by Equation (26). Because the array output is not constant throughout the
day and not all of the solar array power will be used to drive the electrolyzer, the electrolyzer has to be
sized to take advantage of the maximum amount of output power that is available after primary daytime
functions in order for the energy balance to work. Therefore, the Pe used to size the electrolyzer is the
difference between the Pamax and the required Pdl as given by Equation (86).

Pe Pamax − Pdl
= (86)

The thermal control for operating the electrolyzer during the daytime is similar to that for the battery
and can be calculated using Equations (77) to (84). Where the radiator thermal power to be dissipated
would be given by Equation (87). The Sve is given by Equation (88), and the corresponding area of the
electrolyzer enclosure (Aee) is given by Equation (89).

ηe ( Pamax − Pdl )
Pr = (87)

Sve =(1 × 10−4 ) − [( 5 × 10−10 ) Pr ] (88)

2
Aee = 6 ( M e Svb ) 3 (89)

The Mfc is calculated in a similar manner as that for the electrolyzer. To size the Mfc, it was broken down
into a number of components and scaled linearly based on nighttime operational power level (Table 20).
The Mfc is given by Equation (90). It is beyond the scope of this study to investigate the nonlinearity of the

NASA/CR—2020-220368 46
 TABLE 20.—FUEL CELL COMPONENT AND
SCALING FACTOR BREAKDOWN (kg/kW)
Fuel cell stack, Sfc............................................ 2.00
Hydrogen separation tank, Shst ........................0.20
Filters, Sf ......................................................... 0.12
Propellant lines and fittings, Spl.......................0.52
Controller unit, Scu .......................................... 0.16
Wiring, Sw ....................................................... 0.30
Heat exchanger, She ......................................... 1.00
Water pump, Swp.............................................. 0.27
Check valves, Schv ........................................... 0.08
Flow regulators, Sfr.......................................... 0.62
Control valves, Scv ........................................... 0.16
Pressure and temperature sensors, Ss ...............0.07
Flow sensors, Sfs .............................................. 0.06
Hydrogen phase separator, Shps .......................0.07
Hydrogen regulator, Shr ...................................0.05
Oxygen separation tank, Sost ............................0.18
Oxygen regulators, Sor.....................................0.09

individual components scaling factors for an electrochemical stack power level. The process fluid conditions
of pressure, flow rate, pressure drop, and temperature strongly influence these scaling factors and impose a
level of uncertainty.

M=
fc Pnl ( S fc + S hst + S f + S pl + Scu + S w + S he + S wp
(90)
+ Schv + S fr + Scv + S s + S fs + S hps + S hr + Sost + Sor )

Since the scaling factors used to size the electrolyzer (Table 6) and fuel cell (Table 20) are linear, they
are not representative of lower power RFC systems below 500 W. This is due to practical concerns such
as structures and attachments that are utilized to manufacture the system that become a larger fraction of
the total mass producing a nonlinear scaling at lower power levels. The RFC reactant tanks are also sized
in a similar method to the tanks for the ISRU system. The volume of hydrogen and oxygen gas stored is
dependent on the total mass of hydrogen and oxygen that is needed to operate the RFC system throughout
the nighttime. To determine the reactant mass, the flow rate of the reactants is needed to produce the
desired nighttime power.
The flow rate of hydrogen and the subsequent total mass of hydrogen (MH2) needed, can be calculated
from Equation (91) using the total required En from Equation (64) and the ηfc given in Table 14. This
relationship is based on Faraday’s constant (96,485 As/mol or C/mol), the hydrogen charge transfer of
2 electrons per mole, and a theoretical cell operational voltage of 1.48 V since the product, water, is
produced in the liquid state.

s  kg 
(1 + RH2 ) ( En , Wh )  3,600  2.02  En (1 + RH2 )
 h  kmol 
M H2 = (91)
 e −
  C  mol  39, 273η fc
 2 mol H  ( η fc ⋅ 1.48 )  96, 485 mol 1,000 kmol 
 2    

NASA/CR—2020-220368 47
 The mass of oxygen needed (MO2) to react with the hydrogen can be determined from the basic
chemical reaction of combining hydrogen and oxygen to form water, given in Equation (56). Based on
this chemical reaction, 2 mol of hydrogen are required to react with 1 mol of oxygen. Since a mole of
diatomic oxygen gas is 8 times that of a mole of diatomic hydrogen gas, the required oxygen mass is one-
eighth that of the hydrogen as shown in Equation (92). It should be noted for this analysis that a residual
of 20 percent (Rres = 0.2) for both hydrogen and oxygen by mass was added to the total amount required
to account for reactants and water left in the lines and tanks of the system.

MO2 = 8MH2 (92)

As with the tanks for the ISRU oxygen production system, it was assumed that the storage tanks for
both the hydrogen and oxygen gases were spherical. The hydrogen (VH2t) and oxygen (VO2t) tank volumes,
given by Equations (93) and (94), respectively, and the corresponding diameters (dH2t and dO2t), given by
Equations (95) and (96), respectively, are determined from the MH2 and MO2 that need to be stored from
Equations (91) and (92). The tank volumes are based on the gas constants for hydrogen and oxygen,
4,157.2 and 259.84 (J/kg)K, respectively.

4,157.2 Z H 2 M H 2 Tsd
VH 2t = (93)
Pt

259.84 M O2 Tsd
VO2t = (94)
Pt

3VH2t
d H 2t = 2 3 (95)
4π

3VO2t
d O 2t = 2 3 (96)
4π
Using the dH2t and dO2t from Equations (95) and (96), the subsequent tank masses, MH2t and MO2t, can
be calculated using Equations (31) to (34). The variables used in these equations for sizing the hydrogen
and oxygen storage tanks are given in Table 21.
The total Mrfc is given by Equation (97).

M rfc =M e + M fc + M O2 + M H2 + M O2t + M H2t + Rres M O2 + Rres M H2 (97)

TABLE 21.—REGENERATIVE FUEL CELL (RFC)

REACTANT STORAGE TANK SIZING VARIABLES
Variable Symbols Hydrogen storage tank Oxygen storage tank
Pressure, psi (MPa) Pt 2,000 (13.8) 2,000 (13.8)
Factor of safety fs 2.0 2.0
Tank material -- Carbon composite tank Carbon composite tank
Material yield strength, psi (MPa) σy 275,000 (1,900) 275,000 (1,900)

NASA/CR—2020-220368 48
 The array power system using the RFC as the energy storage medium consists of the PV array blanket
and structure, wire, junction box, and RFC system. The mass of these components, as outlined previously,
are summed to provide the total array power system mass (Marfc) as given by Equation (98).

M arfc = M apv + M as + M w + M bb + M rfc + M tc (98)

5.2 Nuclear Kilopower Reactor Power System

Nuclear power systems for space applications have mainly been radioisotope-based systems.
Radioisotope power systems utilize the radioactive decay of an isotope to produce heat through α emissions.
Plutonium is an ideal isotope for this type of application. Isotope-based power systems have been used
extensively since the early days of space exploration with recent use on a number of deep space missions
such as NASA’s Cassini probe and Mars Curiosity rover. However, isotope-based power systems are not
practical for power requirements greater than approximately 1 kW due to the amount of isotope required to
produce the electrical power. Therefore, radioisotope power is not considered a viable option for the ISRU
oxygen production and base camp applications.
As an alternative to a solar-based power source, a nuclear system that utilized fission instead of
radioactive decay would be applicable to the proposed mission. This type of system advantages and
disadvantages for providing the ISRU and base camp power are listed as follows.
Advantages:

• Fission-reactor power systems can scale to very high output power levels and meet the power
demands of the ISRU oxygen production system and base camp.
• The fission power system can operate day and night with no operational constraints.
• The operation of the fission power system is location independent and can be designed (mainly
by adjusting the radiator sizing) to operate anywhere on the lunar surface with little impact on the
system mass.

Disadvantages:

• Fission power systems currently have a low TRL and would need development to be brought up
to flight status.
• The shielding requirements for the reactor can be significant, especially for an inhabited base
camp. Reactor shielding may require infrastructure be available to enable regolith use as a
shielding material, or the reactor may need to be placed a considerable distance away from the
base camp and power loads. This would require the deployment of long cable runs to provide
power from the reactor to the loads.

Since the beginning of space exploration, reactor power systems have always been considered as a
viable option for space power. The first U.S.-based reactor development took place in the 1950s and 1960s
under the Systems for Nuclear Auxiliary Power (SNAP) program (Ref. 25). Under this program, the U.S.
flew its first space nuclear reactor SNAP–10A in 1965, which produced 500 W of electrical power using
thermoelectric (thermal to electric) energy conversion. In conjunction with the SNAP program, a nuclear
reactor for nuclear thermal propulsion was being developed under the Project Rover program. This program
resulted in some ground testing of the Kiwi reactor designs but did not continue on to flight tests. This
transitioned into the Nuclear Engine for Rocket Vehicle Application (NERVA) program in the 1960s, which
continued the ground testing and the development of the nuclear rocket design. The NERVA program did
result in a successful ground demonstration of a 1,000 MWth nuclear rocket engine. Additional

NASA/CR—2020-220368 49
development continued with the Phoebus reactor design, which was the most powerful space reactor
successfully ground tested at 4,000 MWth. Between the Project Rover and NERVA programs, 17 reactors
were developed and tested (Ref. 25).
After the Project Rover and NERVA programs ended in the early 1970s, space reactor development was
halted until 1979 when the Department of Energy (DOE) began the Space Power Advanced Reactor
(SPAR) program development. The SPAR was designed to produce 1,200 kWth power at a temperature of
1,500 K. The SPAR program transitioned to the NASA-led space power reactor program (SP–100) to
develop a 100-kW electric space reactor in the mid-1980s. The SP–100 reactor was capable of producing
2.4 MWth power at a temperature of 1,350 K. By the early 1990s, NASA and DOE began to look into
scaling down the SP–100 reactor to produce between 5- and 20-kW electrical output using either
thermoelectrics or a closed-cycle Brayton power conversion system. The SP–100 program was eventually
canceled in 1994 before any full-scale reactor tests were performed. In 1985, the DOE and Strategic
Defense Initiative Office (SDIO) began a program concurrent with the SP–100 program to construct a larger
multimegawatt space reactor termed the Multimegawatt Space Reactor (MMW) program. This program was
also canceled in 1990 before reactor tests begun. In this same time period, DOE began a program to develop
and test thermionic energy conversion for space applications and constructed a test reactor termed a
Training, Research, Isotopes, General Atomic (TRIGA) reactor. This program was ended in 1993 (Ref. 25).
During the late 1980s and early 1990s, additional space reactor programs were also underway. The
Russians were developing a thermionic reactor called TOPAZ. The first generation of this reactor was
flown and powered two satellites in 1987. The second-generation TOPAZ reactor was designed to
produce a 6-kW electrical output from a 15-kWth heat generation. The Department of Defense under the
SDIO started a program to evaluate the Russian reactors termed the Thermionic System Evaluation Test
(TSRT). Under TSRT, the TOPAZ–II reactor was slated to be used for providing power for a satellite
electric propulsion system. This in turn became the Nuclear Electric Propulsion Space Test Program
(NEPSTP). This would have been the first U.S. launch of a reactor since the SNAP–10A in 1965.
However, the program was canceled in 1996 before the flight tests could take place. In addition, DOE was
looking into a particle bed reactor for space nuclear thermal propulsion applications termed Timberwolf.
It was renamed the Space Nuclear Thermal Propulsion (SNTP) when it was transferred to the Air Force.
This reactor design was different from the design used under the NERVA program for nuclear propulsion.
SNTP was mainly a technology development program and was canceled in 1996 (Ref. 25).
The next significant effort in developing a space nuclear reactor began in 2003 under the Prometheus
program funded by NASA. The main objective for the development of this reactor was the Jupiter Icy
Moons Orbiter (JIMO) mission. The Prometheus reactor was to provide at least 200 kW electrical power
output from over 1 MWth power output. The Prometheus reactor design was new and developed by the
DOE Office of Naval Reactors (DOE–NR). Due to changing priorities within NASA, the program was
canceled in 2005 before conducting any reactor testing (Ref. 25).
The latest and ongoing space reactor development effort is the Kilopower program, which began in
2015 (Ref. 26). This program initially began its development with a 1-kW electrical output space power
reactor based on a 4-kWth power production utilizing Stirling engines for the thermal to electrical energy
conversion. The current development focus is now on a 10-kW electrical output reactor based on a 43-
kWth power output. The reactor is being developed jointly between NASA and DOE’s National Nuclear
Security Administration (NNSA). The reactor utilizes a highly enriched uranium core to generate heat and
uses sodium heat pipes to move the generated heat from the core to eight Stirling engines
(Ref. 27). The Stirling engines then reject their waste heat to a radiator using water heat pipes. To
simplify the reactor system design, the reactor core uses an existing reactor fuel form and the Stirling
engine converters based on the Advanced Stirling Radioisotope Generator (ASRG) design. The reactor
system is illustrated in Figure 39 with a fixed radiator and in Figure 40 with a deployable radiator system.

NASA/CR—2020-220368 50
 Figure 39.—Kilopower reactor system with fixed radiator (Ref. 26).

Figure 40.—Kilopower reactor system with deployable radiator (Ref. 28).

The Kilopower reactor system is intended to provide reliable power for extended operation in space
as well as on planetary surfaces such as the Moon. Multiple units can be “ganged” together to modularly
build up output power capability, as illustrated in Figure 41. To install the Kilopower reactors, it is
envisioned that they would be robotically placed at a safe separation distance from the power loads with a
power transmission cable spooled out as the reactor is moved to its operating location. After placement on
the surface, potentially behind a berm or natural feature to reduce separation distance from the load, any
required radiators are deployed before activating the units. Full power is available in 4 to 6 h, depending
on the size of the reactor. Following shutdown, robotic operations may commence in close proximity to
the Kilopower reactor units within 1 day, or 1 week for crewed operations. This enables relocating the
systems to a new location if desired (Ref. 28).

NASA/CR—2020-220368 51
 It is envisioned that the current electric Kilopower reactor design can be easily scaled from the test,
validated a 1-kW design to a 10-kW electric system with minor changes to the size of the reactor core and
integrating the heat pipes into the core rather than mounting them on the exterior of the core. Table 22
contains the specifications of the 1- and 10-kWe Kilopower reactor systems along with specifications for
a next-generation advanced reactor system which, if developed, would benefit the integrated operation of
the ISRU oxygen production and base camp.

Figure 41.—Kilopower fission-based power system layout.

TABLE 22.—ELECTRIC KILOPOWER REACTOR SPECIFICATIONSa

Parameter 1-kWe reactor 10-kWe reactor 10-kWe core-based
baseline advanced
Nuclear fuel 93 percent enriched solid cast U-235 Mo alloy
Reactor type Fast spectrum, Be reflector, and single centered control rod
Design life >10 yr
Reactor operating temperature, °C 800 800 1,000
Advanced Stirling engine efficiency, percent 23 23 35
Reactor thermal power output, kW 4.3 43.3 43.3
Electrical output power, kW 1 10 10
Output power voltage, Vdc 120 120 120
Distance (dose rate of 26.3 rem/yr), m 100 500 500
Distance (dose rate of 5 rem/yr), km 0.23 1.15 1.15
Reactor radiator area, m2 3.2 20 20
Stowed diameter, m 1.1 1.5 1.5
Stowed height (fixed radiator), m 3.0 7.3 7.3
Stowed height (deployable radiator), m 2.0 3.3 3.3
Reactor mass (Mfr), kg 136 235 235
Shield mass (Mfrs), kg 148 547 547
Balance-of-plant mass (Mfrb), kg 122 763 763
Total mass, kg 406 1,545 1,545
aReference 28.

NASA/CR—2020-220368 52
 Figure 42.—Effect of regolith thickness on radiation dose.

The baseline reactor shielding mass given in Table 22 provides a yearly dose rate for personnel of
26.3 rem at a distance of 500 m for the 10-kWe output reactor. The current guidelines for radiation workers
are a whole-body dose limit of 5 rem/yr (Ref. 29). This is significantly less than values shown in Table 22.
However, this exposure rate can be reduced by increasing the separation distance between the reactor and
the personnel. The dose rate can be scaled based on Equation (99). This equation approximates the reduction
in radiation intensity based on the spherical area as the distance from the reactor changes from the initial
radius (r1) with a known dose (d1) to the desired distance (r2) and subsequent dose (d2).

 r2 
d 2 = d1  12  (99)
 r2 
Using Equation (99), the minimum required distance from the 10-kWe reactor necessary to reduce the
dose rate below the specified 5 rem/yr is 1.15 km. This required distance presumes a direct line of sight
between the reactor and the location where people will be working such as the base camp. Another
approach to reduce this minimum separation distance while maintaining the dose rate limit protection is to
impose a physical barrier. The regolith material itself is a good radiation shielding material and can be
used as a low-cost and mass solution. The dose received through a range of regolith thicknesses from an
unshielded reactor at a distance of 1 km is shown in Figure 42.
Figure 42 shows that the radiation dose drops off significantly with increasing regolith material
thickness. The regolith itself can be built up around the reactor using the Kilopower robotic deployment
systems to move and pile the regolith or by constructing blocks or bags of regolith and stacking them
around the reactor. This approach is illustrated in Figure 43. Using the regolith to shield the reactor in this
manner has the following requirements and limitations:

• Access to the equipment is needed to pile the regolith to a sufficient height and thickness.
• The area above the reactor will still be exposed when required to be left open. If it can be closed,
a fixed shield will need to be placed over the reactor. Another option would be to place a
container over the reactor to support and pile the regolith on top of the container to provide
complete shielding of the reactor.

NASA/CR—2020-220368 53
 • Variations of the mounded regolith shielding can be considered:
○ Partial shielding, where only a certain area around the reactor is shielded; the unshielded
areas would be off limits to personnel.
○ Utilizing a natural feature such as an outcrop as part of the shielding. In this situation, the
reactor will be placed near an outcrop and the regolith piled up around the exposed areas of
the reactor. This will reduce the amount of regolith required to be moved.
○ A final variation for utilizing the regolith directly as a shielding material is to change its form
by melting the regolith into blocks or filling up bags with the regolith. The blocks or bags can
then be stacked to form a radiation shield. The blocks or bags may be easier to handle and
provide a more accurate shielding design.

Figure 43.—Reactor shielding using loose regolith.

NASA/CR—2020-220368 54
 Figure 44.—Reactor shielding by burying reactor.

Another shielding option is to bury the reactor by digging a pit and placing the reactor in it. This
approach is illustrated in Figure 44. Placing the reactor in a constructed pit can provide a significant
amount of shielding. The following are some concerns or considerations for this approach:

• The pit dimensions, particularly the depth, will determine the feasibility of this type of shielding.
The depth requirement will be set by the location of the reactor core and the desired exclusion
range from the pit.
• Excavation equipment needed to prepare the pit will require an alternate power source before the
reactor is operational.

NASA/CR—2020-220368 55
 • Neither the regolith depth nor density is uniform over the surface of the Moon. This may limit the
reactor location to areas where there is sufficient depth of lunar regolith with adequate
consistency to place the pit.
• The ability to place the reactor into the pit and provide all the necessary thermal and electrical
connections to components on the surface.
• Since the reactor will be below grade, there may be a need to provide shielding over the pit
depending on activity within and above the area where the reactor is located. Options for the pit
top include
○ Maintain a keep-out zone above the reactor site
○ Utilize a partial shield around the perimeter of the pit
○ To place a cover over the reactor pit and utilize loose regolith to shield the top of the pit

The last option is to utilize the terrain for shielding the reactor. A crater or depression in the terrain
can be used to provide shielding as illustrated in Figure 45. This is the least intrusive method and does not
require any equipment to prepare the site. However, it does limit the potential locations for the reactor as
well as the mission and equipment supported by the reactor power. Some considerations for this type of
shielding include

• A crater with sufficient depth is needed to meet the shielding requirements at or near the location
where the proposed mission is to take place.
• The ability to place the reactor into the crater and provide all of the necessary thermal and
electrical connections is needed.
• The crater width and therefore wall steepness will affect the ability to access the reactor as well as
the need for additional shielding on the reactor.
• As with the pit shielding, the area above the crater will need to be a keep-out zone for personnel.
• Additional shielding can be applied to reduce or eliminate any operating concerns with activity
taking place over the crater. This can be accomplished by placing a cone shield on top of the
reactor to enable personnel to move to the crater edge or over the crater.

Figure 45.—Reactor terrain shielding by placing it in crater.

NASA/CR—2020-220368 56
 Figure 46.—Distance to horizon versus height above
lunar surface.

Similar to the terrain shielding by using a crater to shield the reactor, another approach would be to
place it over the horizon from where the base camp and personnel will be. This would require a cable to
be deployed from the reactor to the power loads, so distance is a considerable factor. However, due to the
smaller lunar radius, this distance is not as significant as it would be on Earth. The distance to the horizon
(dhr) for a given height above the lunar surface (h) is given by Equation (100) and plotted in Figure 46.

d hr = h ( 2rm + h ) (100)

If the h of the personnel or components requiring protection is on the order of 3 m or less, the horizon
distance is approximately 3 km. This distance is greater than the current required distance with the
baseline shielding to stay below the specified 5 rem/yr. However, this approach does provide some unique
benefits. It could potentially reduce the required shielding mass on the reactor since the large amount of
regolith between the personnel and the reactor would provide a more than adequate amount of shielding
to reduce the radiation exposure to background levels.
If personnel are required to be in close proximity to the power loads, such as the base camp, then the
reactor will need to be placed sufficiently far enough away to keep below the required maximum
personnel dose level. As given in Table 22, this would require the reactor to be placed 1.15 km from the
base camp and/or oxygen production plant location for the 5-rem/yr dose level with the 10-kWe reactor.
However, if multiple reactor systems are needed, this distance will increase. For three 10-kWe reactors, it
would be 1.99 km and for six it would be 2.81 km. At this latter distance, the reactors will be over the
horizon for most personnel standing on the surface.
To achieve this distancing, a cable will need to be deployed that connects the reactor to the power
loads. Also, since the reactor has an output of 120 Vdc, electrical converters at both the reactor and power
loads will be needed to raise the output power voltage for transmission and then reduce it at the loads for
use. This is done to reduce the overall required cable wire gauge and subsequent mass. The reactor Mw is
calculated in a similar manner to the array Mw. The Iw the reactor wire needs to carry is based on the total

NASA/CR—2020-220368 57
power requirement of the loads. For the reactor system, the required power is continuous both day and
night. This is similar to the continuous daytime power level used in the array system analysis. The
number of wires used to transmit the power is based on the number of 10-kWe reactor systems needed to
meet the power system requirements. The reactor power systems are modular systems providing up to 10
kWe of power each. The number of reactor systems needed (nrs) to meet the power demands of the
various mission options is given by Equation (101) where the value is rounded up to the next whole
number to provide whole reactor units. Subsequently, the required Iw per transmission wire is determined
by the voltage of the power wires (Vrt) as given by Equation (102).

 P + Pwl 1 
=nrs  dl + (101)
 10,000 2 

Pdl
Iw = (102)
Vrt nrs

With the Iw calculated from Equation (102), the Gw and subsequent Mw can then be calculated using
Equations (71) and (73), respectively. For the reactor system, the values for the variables used in the wire
mass calculation are given in Table 23 and the total nw used is given by Equation (103).

nw = nrs nrw (103)

Due to the Lw between the reactor and the loads, the power transmission loss down the wire (Pwl) also
needs to be considered. The resistance of a copper wire per meter of length (Rw), as a function of its Gw, is
given by Equation (104) and the corresponding Pwl is given by Equation (105).

=Rw 0.00032 × 100.23188Gw (104)

Pwl = I w2 Rw Lw (105)

To minimize the transmission line mass, the voltage was boosted from the 120 Vdc output of the fission
reactor to 1,500 Vdc for transmission. Each transmission power line needs a direct current (dc) to dc
converter at the reactor to raise the voltage from 120 Vdc to 1,500 Vdc and then one at each power load to
reduce the voltage from 1,500 Vdc back to 120 Vdc. To scale the converter mass (Mvc), data was collected
from operating the Mvc over a range of power levels and over a similar conversion voltage range as that
needed for the fission reactor power transmission as shown in Figure 47. The data was then curve fit to
provide an expression to calculate the Mvc as a function of the operating power level, as given by Equation
(106).

Pdl
M vc =
  Pdl   (106)
1,000nrs 0.0279   + 0.6816 
  nrs 1,000  

TABLE 23.—FISSION REACTOR POWER

TRANSMISSION WIRE SPECIFICATIONS
Number of wires per reactor system, nrw ........................2
Transmission voltage, Vrt, V ................................... 1,500
Transmission line length, Lw, m .............................. 1,150

NASA/CR—2020-220368 58
 Figure 47.—Direct current (dc) to dc converter specific
power versus operating power level.

The total reactor system mass (Mfrt) is a sum of the reactor components plus the wiring, given by
Equation (73), and conditioning. The Mfrt is given by Equation (107).

M frt= nrs M fr + nrs M frs + nrs M frb + M w + 2nrs M vc (107)

6.0 Results
Using the analyses described in the previous sections, results were generated for three cases. Each of
the cases compares the mass of the PV array-based power system to that of the Kilopower fission reactor-
based system. The first two cases look at the power system sizing for the ISRU oxygen production plant
and base camp independently. Whereas the third case combines the power requirements for both
applications. It should be noted that the reactor systems are not scaled to directly match the electrical and
thermal requirements of the proposed systems but are utilized as building blocks with a fixed output that
can be tied together to operate as a single power system if the required power exceeds the output
capability of one reactor system. This modular-type power system approach is beneficial in terms of unit
cost since the reactor power systems are identical in construction. It can provide excess power capability
to a power load beyond the baseline power needs. In that instance, there will be a mass penalty associated
with the reactor system over a custom-designed system for the application.

6.1 Case 1—Power System Comparison for ISRU Oxygen Production Plant
The baseline oxygen production rate was selected as 1.63 kg/h. At this rate, over 1 lunar day and
night cycle, 1,154 kg of oxygen can be produced. The thermal and electrical power requirements for this
baseline oxygen production level are given in Figure 11 and Figure 12 and the mass and power
requirements for the oxygen production system are summarized in Table 24.

NASA/CR—2020-220368 59
 To meet the power requirements for the ISRU oxygen production system, the solar PV array power
system would provide electrical power for both the electrical and thermal requirements of the system. For
continuous operation, as discussed in Section 5.1 and as shown in the energy balance diagram of the PV
array and battery power system in Figure 48, the array would provide sufficient power during the daytime
to operate the oxygen production system as well as recharge the energy storage system to enable ISRU
operation during the nighttime.

TABLE 24.—BASELINE OXYGEN PRODUCTION MASS AND

POWER REQUIREMENTS AT PRODUCTION RATE OF 1.63 kg/h
Component Required power, Mass,
W kg
Scoop and delivery conveyors 134 51
Vibrating screen 132 68
Magnetic separator 98 98
Fluidized bed reactor --------- 27
Hydrogen tank --------- 19
Oxygen tank --------- 488
Electrolyzer 8,965 49
Thermal power to process reactor 13,386 -------
Heat of reaction 2,563 -------
Heat loss 552 -------
Thermal control system --------- 326
Total 25,830 1,125

Figure 48.—Photovoltaic (PV) array and battery power system’s energy balance for day
and night oxygen production at 25.8 kW of continuous output power. Required array
area is 208.5 m2. Total energy required for 1 day and night cycle is 18,288 kWh. The
array is at 30° latitude with a fixed 30° tilt angle and 28 percent efficient solar cells.
Earth spring equinox is 1,359 W/m2.

NASA/CR—2020-220368 60
 The PV array power system can use either a battery or RFC for energy storage. The mass breakdown of
the PV array and battery power system to meet the power and energy requirements are shown in Figure 49.
The energy balance for the PV array and RFC system is given in Figure 50 and the corresponding system
mass breakdown is given in Figure 51. The energy balance differs between the PV array and battery and the
PV array and RFC systems due to their difference in charging and discharging efficiencies. Because of the
lower efficiency of the RFC charge and discharge cycle, the PV array area increases to 353.6 m2 from the
248.5 m2 required by the battery system. In addition, the corresponding peak operating output power of the
PV array increased from 84.4 kW for the battery system to 120.2 kW for the RFC system.

Figure 49.—Photovoltaic (PV) array and battery system mass breakdown for continuous oxygen
production at 25.8 kW of continuous output power.

Figure 50.—Photovoltaic (PV) array and regenerative fuel cell (RFC) power system
energy balance for day and night oxygen production at 25.8 kW of continuous output
power. Required array area is 353.6 m2. Total energy required for 1 day and night
cycle is 18,288 kWh. The array is at 30° latitude with a fixed 30° tilt angle and
28 percent efficient solar cells. Earth spring equinox is 1,359 W/m2.

NASA/CR—2020-220368 61
 Figure 51.—Photovoltaic (PV) array and regenerative fuel cell (RFC) system mass breakdown for
continuous oxygen production at 25.8 kW of continuous output power.

The results show that for continuous operation of the PV array-based power system, the energy storage
mass dominates the total mass of the system, 95 percent for a battery and 75 percent for a RFC. The battery
specific energy was given as 200 Wh/kg with an 80 percent depth of discharge while the RFC specific
energy was 832 Wh/kg based on the mass scaling analysis given in Section 5.1 and the required operating
time and power level set by the energy balance curve given in Figure 50. Although the energy storage mass
for both the battery and RFC was the dominant mass in their respective power systems, their comparison
illustrates the advantage the RFC system has over batteries for long-term operation. For a fixed output
power, as in this case, a longer operating time results in a larger energy storage requirement. This causes the
specific energy of the RFC system to increase whereas the battery has a constant specific energy regardless
of operating time. The reason the RFC system specific energy increases with longer operational time is
because all of the components except for the reactants and storage tanks are sized based on the maximum
output power level. Only the tanks and reactants scale or increase with longer operation. Therefore,
operating for longer duration does not increase the component masses, only that of the tanks and reactants
required to support the larger energy storage requirement. This results in the specific energy of the system
increasing with increasing operating time at a fixed power level.
Because the nighttime energy storage requirements are such a dominating factor in the sizing of the PV
power system, another approach can be taken for powering the ISRU oxygen production system. Oxygen
can be produced only during the day and the system can be placed in a hibernation mode during the night
utilizing only enough power to maintain the equipment at a temperature sufficient for it to survive the night.
However, in order to accurately compare this approach to the baseline case of producing 1,154 kg of oxygen
over 1 day and night cycle, the system will need to produce this much while operating just during the
daytime. Therefore, the oxygen production system has to be scaled up to operate at a higher production rate
during the daytime operation. The production rate needed to achieve a similar total mass of oxygen while
operating just during the daytime over the period of 218.3 h identified in Figure 52 is 5.29 kg/h, which
produces a total oxygen mass of 1,158 kg. The mass and power breakdown of the oxygen production system
at this production rate is given in Table 25. The energy balance shown in Figure 52 for the PV array and
battery system describes this approach. The nighttime power requirement will be the power needed to
maintain the equipment at a temperature that will allow it to survive the nighttime. This is calculated in a
similar manner as the nighttime heat loss for the habitat as given in Section 4.0. Based on this analysis the
nighttime heater requirement for the oxygen production system was estimated to be approximately 500 W.

NASA/CR—2020-220368 62
 Figure 52.—Photovoltaic (PV) array power system energy balance for daytime-only
oxygen production at 82.7 kW daytime operational and 500 W nighttime “keep-alive”
power. Required array area is 327 m2. Total energy required for 1 day and night cycle is
18,288 kWh. The array is at 30° latitude with a fixed 30° tilt angle and 28 percent
efficient solar cells. Earth spring equinox is 1,359 W/m2.

TABLE 25.—DAYTIME-ONLY OXYGEN PRODUCTION MASS AND

POWER REQUIREMENTS AT PRODUCTION RATE OF 5.29 kg/h
Component Required power, Mass,
W kg
Scoop and delivery conveyors 160 51
Vibrating screen 430 91
Magnetic separator 317 98
Fluidized bed reactor NA 27
Hydrogen tank NA 19
Oxygen tank NA 488
Electrolyzer 29,095 159
Thermal power to process reactor 43,446 NA
Heat of reaction 8,316 NA
Heat loss 552 NA
Thermal control system NA 1,059
Total 82,315 1,989

NASA/CR—2020-220368 63
 The daytime-only operational breakdowns of the PV array and battery and the PV array and RFC
power system masses are shown in Figure 53 and Figure 54, respectively. The mass breakdown for the
RFC system is given in Table 26. The total mass for the PV array systems also includes the increased
mass of the oxygen production system needed to produce oxygen at the higher rate to match the day and
night continuous operation production amount. The daytime-only operation has significantly reduced the
overall system mass for both systems over the continuous operation systems. However, the energy storage
mass for both systems, even though it is used only for nighttime keep-alive power, is still a significant
item for both systems. For the battery system, it is the largest mass item, and for the RFC system, it is the
second largest item behind the array.

Figure 53.—Photovoltaic (PV) array and battery system mass breakdown for daytime oxygen production for
82.7 kW daytime operational and 500 W nighttime “keep-alive” power.

Figure 54.—Photovoltaic (PV) array and regenerative fuel cell (RFC) system mass breakdown for daytime
oxygen production for 82.7 kW daytime operational and 500 W nighttime “keep-alive” power.

NASA/CR—2020-220368 64
 TABLE 26.—REGENERATIVE FUEL CELL (RFC) SYSTEM
MASS BREAKDOWN FOR NIGHTTIME “KEEP-ALIVE” POWER
Fuel cell Electrolyzer
Fuel cell stack, Sfc ............................................. 1.000 Electrolyzer stack, Ses ........................................ 56.9
Hydrogen separation tank, Shst .......................... 0.100 Water tank and heater .......................................... 2.9
Filters, Sf ........................................................... 0.060 Filters, Sf .............................................................. 3.4
Propellant lines and fittings, Spl......................... 0.260 Propellant lines and fittings, Spl ......................... 14.8
Controller unit, Scu ............................................ 0.080 Controller unit, Scu ............................................... 4.6
Wiring, Sw ......................................................... 0.150 Wiring, Sw ............................................................ 8.5
Heat exchanger, She ........................................... 0.500 Heat exchanger, She............................................ 28.5
Water pump, Swp ............................................... 0.135 Water pump, Swp .................................................. 7.7
Check valves, Schv ............................................. 0.040 Check valves, Schv ................................................ 2.3
Flow regulators, Sfr ........................................... 0.310 Flow regulators, Sfr ............................................ 17.5
Control valves, Scv............................................. 0.080 Control valves, Scv ............................................... 4.6
Pressure and temperature sensors, Ss................. 0.040 Pressure and temperature sensors, Ss ................... 2.0
Flow sensors, Sfs................................................ 0.030 Flow sensors, Sfs .................................................. 1.7
Hydrogen phase separator, Shps ......................... 0.040
Total ................................................................ 155.3
Hydrogen regulator, Shr ..................................... 0.030
Oxygen separation tank, Sost.............................. 0.090 System total ..................................................... 309.4
Oxygen regulators, Sor ...................................... 0.050
Hydrogen tank ................................................ 30.700
Hydrogen and residual .................................... 12.000
Oxygen tank .................................................... 12.500
Oxygen and residual ....................................... 95.900
Total .............................................................. 154.100

The RFC system specific energy for providing 500 W of electrical power throughout the nighttime
with the mass breakdown given in Table 26 was 456 Wh/kg. This is lower than the RFC specific energy
calculated for continuous operation due to the smaller total energy storage requirement resulting from
operating at only 500 W. In general, as the operating power decreases, the RFC specific power also
decreases. This decrease is due to the nonlinear scaling of some of the components at lower operating
power levels as well as the effect of the reactants becoming a smaller fraction while the electrical
conversion components become a higher fraction of the total RFC mass.
The reactor system was also sized to meet the ISRU oxygen production requirements. For the reactor
system, two different approaches were analyzed. One in which the reactor was located near the oxygen
production facility, and the second where it was located a distance away, 1.99 km, to maintain the
maximum 5 rem/yr dose rate for personnel at the oxygen production facility location.
For the reactor located near the oxygen production facility, it is assumed that the facility is fully
automated and that no personnel are needed for its operation. If crew interaction with the facility is
required to remove the oxygen produced or for servicing the oxygen production facility, the reactor will
need to be shut down during that time to maintain a safe radiation environment for any personnel within
the exclusion zone. With the reactor located near the oxygen production facility, it would be possible to
utilize the waste heat from the reactor to provide heat for the oxygen production process. The baseline
production-rate thermal and electrical power requirements from Figure 11 and Figure 12, respectively, are
16,502 and 9,329 W.
The baseline 10-kWe reactor system can provide the electrical output requirement for the ISRU system
but not the thermal requirement. However, if two reactor systems are utilized, there will be sufficient
electrical power and heat available to meet the oxygen production system requirements. The heat produced

NASA/CR—2020-220368 65
 Figure 55.—Baseline reactor system mass and power for the oxygen production plant using waste heat
from the reactor for continuous oxygen production.

TABLE 27.—BASELINE REACTOR SYSTEM PARAMETERS

Item Value
Heat output per reactor at 800 °C, W .................................... 43,300
Reactor mass, each, kg ............................................................... 235
Shielding per reactor, kg ............................................................ 547
Balance of plant per reactor, kg ................................................. 763
Number of reactors ........................................................................ 2
Total mass for two reactor power systems, kg ........................ 3,090
Electrical power required, W .................................................. 9,329
Electrical heat needed to raise regolith to 1,000 °C, W ........... 3,300

by the baseline reactor system is at 800 °C. Therefore, electrical heaters will still be needed to add additional
heat to 1,000 °C to meet the requirements of the oxygen production system. The breakdown of the heat
output of the reactors and the mass of the system is given in Figure 55 and Table 27.
As discussed previously, the reactors are modular power systems that provide fixed output, both
thermal and electrical. For Case 1, two reactor systems were able to meet the thermal and electrical
requirements of the oxygen production system. However, the reactors also produced an excess 18,488 W
of heat. This unused capability will increase the overall mass of the reactor system beyond that of a
reactor custom sized for this application.
To better match the reactor to the oxygen production system requirements, an advanced system that
operates at a higher temperature and greater Stirling engine efficiency can be conceived. This system, as
described in Table 22, would be better matched to the oxygen production system. By utilizing the heat
generated by the reactor directly in the oxygen production reactor, only one fission reactor system would
be required. The thermal and mass breakdown for this reactor system is given in Figure 56 and Table 28.
The last option for the reactor system is one in which the reactors are located a distance from the
oxygen production system in order to maintain a dose level below the 5 rem/yr limit. Due to the distance
away from the oxygen production system, thermal heat from the reactor cannot be directly utilized.
Therefore, the reactor will need to provide electrical power to meet both the thermal and electrical power
requirements of 25,830 W for the oxygen production system as presented by Figure 13. To accomplish
this, three 10-kWe fission-power reactor systems will be needed. The dose from the three-reactor system
causes the required distance to increase from the 1.15 km for one reactor to 1.99 km. The mass
breakdown for this power system is shown in Figure 57.

NASA/CR—2020-220368 66
 Figure 56.—Advanced reactor system mass and power for the oxygen production plant using waste
heat from the reactor.

TABLE 28.—ADVANCED REACTOR SYSTEM PARAMETERS

Item Value
Heat output at 1,000 °C, W ................................................... 43,300
Thermal power required at 1,000 °C, W ............................... 16,502
Total heat to the Stirling engines for electrical power, W ..... 26,653
Electrical power required, W .................................................. 9,329
Reactor mass, kg ........................................................................ 235
Shielding, kg .............................................................................. 547
Balance of plant, kg ................................................................... 763
Total mass, kg ......................................................................... 1,545

Figure 57.—Baseline three-reactor system mass breakdown for continuous oxygen

production at a distance of 1.99 km from the plant. Direct current is dc.

NASA/CR—2020-220368 67
 Figure 58.—Power system mass comparison for regolith oxygen production.

The overall power system mass comparison for powering the ISRU oxygen production system is
shown in Figure 58. The continuous operation PV array system utilizing batteries for energy storage is by
far the largest mass system. Ignoring the continuous operation PV array systems, the remaining systems
and operational approaches are all comparable in mass with the advanced fission reactor approach, which
has the overall lowest mass. The PV array and RFC system operating only during the daytime has the
lowest mass of the PV systems and is lower than that of the currently designed reactor systems.

6.2 Case 2—Base Camp’s Power System

Unlike the oxygen production system, the base camp’s power system is required to operate continuously
day and night at the same power level. Therefore, the PV system for this application must have sufficient
storage to operate throughout the night. The energy balance to meet the habitat’s continuous day and night
power requirement of 28.05 kW, as discussed in Section 4.4 and summarized in Figure 26, is given in
Figure 59 for the PV array and battery system. The mass breakdown for the PV array and battery system is
given in Figure 60. The battery mass for this system constitutes approximately 95 percent of the total system
mass. This is due to the requirement of operating at full power throughout the night.
The energy balance for the PV array and RFC system is given in Figure 61. Based on this energy
balance, the power system mass was determined as shown in Figure 62. As seen in Case 1, the RFC
provides a lower mass system for energy storage for operation throughout the lunar night than the battery
for the base camp operation. However, the RFC still constitutes approximately 92 percent of the total
power system mass. The breakdown of the RFC system mass is shown in Table 29. The RFC system has
a total specific energy of 831 Wh/kg. The array size was larger for the RFC than the battery system due to
the lower charge and discharge efficiency of the RFC. This can be seen in the energy balance diagram
for the PV array and RFC system (Figure 61). The required array size for the PV and RFC system is
42 percent larger than the array area needed for the PV and battery system.

NASA/CR—2020-220368 68
 For the fission reactor power system to provide adequate shielding for the personnel at the habitat for
a maximum dose rate of 5 rem/yr, the reactors were positioned 1.99 km from the habitat. This is similar to
the approach used with the crew-tended oxygen production in Case 1. As discussed in Section 5.2,
material shielding could also be employed to reduce the separation distance between the reactor and the
habitat. However, for this analysis, the 1.99 km distance was utilized with no additional terrain shielding
considered.

Figure 59.—Photovoltaic (PV) array and battery energy balance for day and night base
camp power at 28.05 kW of continuous output power. Required array area is 226.5 m2.
Total energy required for 1 day and night cycle is 19,869 kWh. The array is at 30°
latitude with a fixed 30° tilt angle and 28 percent efficient solar cells. Earth spring
equinox is 1,359 W/m2.

Figure 60.—Lunar base camp photovoltaic (PV) array and battery mass breakdown at 28.05 kW
of continuous output power.

NASA/CR—2020-220368 69
 TABLE 29.—REGENERATIVE FUEL CELL (RFC) SYSTEM MASS BREAKDOWN FOR BASE CAMP POWER
Fuel cell Electrolyzer
Fuel cell stack, Sfc ........................................... 56.100 Electrolyzer stack, Ses ........................................205.0
Hydrogen separation tank, Shst .............................. 5.6 Water tank and heater ..........................................10.3
Filters, Sf ............................................................... 3.4 Filters, Sf ..............................................................12.3
Propellant lines and fittings, Spl .......................... 14.6 Propellant lines and fittings, Spl ...........................53.3
Controller unit, Scu ................................................ 4.5 Controller unit, Scu ...............................................16.4
Wiring, Sw ............................................................. 8.4 Wiring, Sw ............................................................30.8
Heat exchanger, She ............................................. 28.1 Heat exchanger, She ............................................102.5
Water pump, Swp ................................................... 7.6 Water pump, Swp ..................................................27.7
Check valves, Schv ................................................. 2.2 Check valves, Schv ................................................. 8.2
Flow regulators, Sfr ............................................. 17.3 Flow regulators, Sfr ..............................................63.1
Control valves, Scv................................................. 4.5 Control valves, Scv................................................16.4
Pressure and temperature sensors, Ss..................... 2.0 Pressure and temperature sensors, Ss..................... 7.2
Flow sensors, Sfs.................................................... 1.7 Flow sensors, Sfs.................................................... 6.2
Hydrogen phase separator, Shps ............................. 2.0 Total ...................................................................559.2
Hydrogen regulator, Shr ......................................... 1.4
System total .................................................... 9,627.3
Oxygen separation tank, Sost ................................. 5.1
Oxygen regulators, Sor .......................................... 2.5
Hydrogen tank ............................................... 1,809.1
Hydrogen and residual ...................................... 706.1
Oxygen tank ...................................................... 737.1
Oxygen and residual ...................................... 5,649.0
Total ............................................................... 9,068.1

Figure 61.—Photovoltaic (PV) array and regenerative fuel cell (RFC) energy balance
for day and night base camp power at 28.05 kW of continuous output power.
Required array area is 384.2 m2. Total energy required for 1 day and night cycle is
19,868 kWh. The array is at 30° latitude with a fixed 30° tilt angle and 28 percent
efficient solar cells. Earth spring equinox is 1,359 W/m2.

NASA/CR—2020-220368 70
 Figure 62.—Lunar base camp photovoltaic (PV) array and regenerative fuel cell (RFC) mass
breakdown at 28.05 kW of continuous output power.

Figure 63.—Power system mass comparison for

providing base camp power.

A power cable would be deployed between the reactor and habitat. An exclusion zone of 1.99 km
surrounding the reactor would be maintained when it is operational. To meet the 28.05 kWe power
requirement for the habitat, three 10-kWe reactors were utilized. The mass breakdown for this system is
the same as the reactor power system shown in Figure 57 for the ISRU oxygen production option.
The comparison of the power systems’ masses for the base camp’s power requirements is given in
Figure 63. As seen in the figure, the Kilopower-based fission reactor system provides the lowest mass for
the base camp’s power system operation. The PV array and battery system is by far the heaviest due to the
large battery mass needed to operate throughout the night. The PV array and RFC system had
significantly less mass, approximately 80 percent, than the battery system, but was still 2.5 times as
massive as the reactor system. This application demonstrates that for lunar missions requiring high power
levels throughout the nighttime period, the reactor system provides a significant mass advantage over PV
array-based systems that need to store energy for nighttime operation.

NASA/CR—2020-220368 71
6.3 Case 3—Oxygen Production and Base Camp Combined Power System
The final case examines power systems operating both the ISRU oxygen production system and the
base camp. For the combined operation, it was assumed that both components were at the same location.
For the reactor power system, this eliminated the option of using the reactor waste heat for the oxygen
production system. For the PV array-based systems, only the operation of the oxygen production during
the daytime was considered, since operating the oxygen production system using stored energy
throughout the nighttime was shown to be noncompletive in Case 1. For the PV array and battery power
system, the daytime power consumption includes both operating the ISRU oxygen production plant as
well as providing power to the base camp. The excess power from the array, above what is used for
oxygen production and the base camp, is used to recharge the energy storage system. The corresponding
energy balance for the battery system is shown in Figure 64.
From the energy balance diagram (Figure 64), the corresponding operating time for the oxygen
production plant is 218.3 h. This requires an oxygen production rate of 5.29 kg/h to meet the 1,154 kg of
oxygen production for 1 day and night period. The mass breakdown for the oxygen production system to
meet this required production rate is given in Table 30.
The total power system mass breakdown for the PV array and battery power system used to power the
combined oxygen production system and base camp is shown in Figure 65. The battery mass still
dominates the overall power system mass. Because the battery mass is such a large fraction of the total
mass, there is not much difference in the overall mass of this system compared to the PV array and battery
system used to power the base camp alone.

Figure 64.—Photovoltaic (PV) array and battery energy balance for base camp and
oxygen production for 172.5 kW daytime and 28.05 kW nighttime operational power.
Required array area is 687 m2. The array is at 30° latitude with a fixed 30° tilt angle
and 28 percent efficient solar cells. Earth spring equinox is 1,359 W/m2.

NASA/CR—2020-220368 72
 TABLE 30.—DAYTIME-ONLY OXYGEN PRODUCTION MASS AND
POWER REQUIREMENTS FOR COMBINED ISRU PLANT AND
BASE CAMP POWER SYSTEM WITH BATTERY ENERGY
STORAGE AT PRODUCTION RATE OF 5.29 kg/ha
Component Required power, Mass,
W kg
Scoop and delivery conveyors 160 51
Vibrating screen 430 91
Magnetic separator 317 98
Fluidized bed reactor --------- 27
Hydrogen tank --------- 19
Oxygen tank --------- 486
Electrolyzer 29,095 159
Thermal power to process reactor 43,445 -------
Heat of reaction 8,316 -------
Heat loss 552 -------
Thermal control system --------- 1,059
Total 82,315 1,990
aIn situ resource utilization (ISRU).

Figure 65.—Photovoltaic (PV) array and battery power system mass breakdown for operating combined
in situ resource utilization (ISRU) oxygen production system and base camp for 172.5 kW daytime and
28.05 kW nighttime operational power.

Because of the difference in their charge and discharge efficiencies, the RFC-based energy storage
system will have a different energy balance than the battery-based system. The energy balance for the PV
array and RFC system to meet the day and nighttime power requirements is shown in Figure 66. The PV
array area for the RFC-based system has increased by approximately 50 m2. The larger array area is
needed for the RFC-based system due to its lower charge and discharge efficiencies compared to the
battery-based system. However, the larger array does provide some benefits enabling the power system to
be operated at a slightly lower peak power level than the PV array and battery system during the daytime.
This is due to having sufficient power provided earlier in the morning and later in the evening to operate
the oxygen production system. This in turn slightly reduces the required oxygen production rate to 4.66
kg/h and thereby lowers the total power requirement over the daytime operating time. It should be noted

NASA/CR—2020-220368 73
that the total energy produced by the system for oxygen production is the same as for the other cases in
order to make the desired 1,154 kg of oxygen over the 1-day period. The mass breakdown for the oxygen
production system to meet this production rate requirement is given in Table 31.

Figure 66.—Photovoltaic (PV) array and regenerative fuel cell (RFC) energy balance for base
camp and oxygen production for 152.2 kW daytime and 28.05 kW nighttime operational
power. Required array area is 730 m2. Total energy required for 1 day and night cycle is
38,158 kWh. The array is at 30° latitude with a fixed 30° tilt angle and 28 percent efficient
solar cells. Earth spring equinox is 1,359 W/m2.

TABLE 31.—DAYTIME-ONLY OXYGEN PRODUCTION MASS AND

POWER REQUIREMENTS FOR COMBINED ISRUa PLANT AND
BASE CAMP POWER SYSTEM WITH RFCb ENERGY
STORAGE AT PRODUCTION RATE OF 4.66 kg/h
Component Required power, Mass,
W kg
Scoop and delivery conveyors 155 51
Vibrating screen 378 89
Magnetic separator 279 98
Fluidized bed reactor -------- 27
Hydrogen tank -------- 19
Oxygen tank -------- 486
Electrolyzer 25,630 140
Thermal power to process reactor 46,150 -------
Heat of reaction 7,326 -------
Heat loss 552 -------
Thermal control system -------- 933
Total 72,593 1,843
aInsitu resource utilization (ISRU).
bRegenerative fuel cell (RFC).

NASA/CR—2020-220368 74
 Figure 67.—Photovoltaic (PV) array and regenerative fuel cell (RFC) power system mass breakdown
for operating combined in situ resource utilization (ISRU) oxygen production system and base camp
for 152.2 kW daytime and 28.05 kW nighttime operational power.

Figure 67 shows the corresponding mass breakdown for the PV array and RFC system to meet the
requirements to power the base camp and oxygen production system per the energy balance diagram
(Figure 66). The PV array and RFC system provides a significant mass reduction over the solar-powered
PV system utilizing batteries for energy storage. The RFC-based system provided a 75-percent decrease
in the required mass over the battery-based energy storage system.
The final power system option to consider for powering the base camp and oxygen production system
is the Kilopower reactor. With the baseline assumption that the oxygen production system is in close
proximity to the base camp, the ability to use the waste heat from the Kilopower reactor as process heat
with the oxygen production system is not an option. Therefore, the only option is to provide electrical
power for both the habitat and the oxygen production system. The general layout of the reactor system,
base camp, and oxygen production system is illustrated in Figure 68.
Since personnel will be working in the vicinity of the base camp and oxygen production system, the
reactor was positioned 2.81 km from the habitat to provide adequate shielding as discussed in the
previous two cases. A power cable would then be deployed between the reactor and habitat. An exclusion
zone of 2.81 km surrounding the reactor would be maintained when it is operational. To meet the
53.84 kWe total power requirement for the habitat and oxygen production system, six 10-kWe reactor
systems can be utilized as illustrated in Figure 51. However, this arrangement would provide an excess
power capacity of 6.2 kW. The mass breakdown for the six-reactor system is given in Figure 69.
The baseline analysis utilizes the reactor systems as a fixed 10-kWe modular power system block that
can be added together to meet the power requirements in increments of 10 kW. However, if the systems
were optimized to provide a power output that better matched the power requirements, then a mass
savings could potentially be realized. Therefore, an additional reactor system analysis was added in which
the reactor mass was scaled linearly to match the total power requirement. The mass breakdown for this
scaled case is shown in Figure 70.

NASA/CR—2020-220368 75
 Figure 68.—Kilopower reactor-based power system, base camp, and oxygen production
system layout. In situ resource utilization is ISRU.

Figure 69.—Baseline reactor system mass breakdown for base camp and oxygen
production at distance of 1.15 km from plant for 53.84 kW continuous day and night
power production. Direct current is dc.

NASA/CR—2020-220368 76
 Figure 70.—Scaled reactor system mass breakdown for base camp and oxygen production
at distance of 1.15 km from plant for 53.84 kW continuous day and night power
production.

Figure 71.—Power system mass comparison for providing base camp and oxygen production system power.

The comparison of the power system mass for power requirements of the combined base camp and
oxygen production system is given in Figure 71. As seen in this figure, the Kilopower-based fission
reactor system provides the lowest mass for the base camp and oxygen production power system. To
reduce the energy storage mass requirements for the PV array-based systems, the oxygen production
system operated only during the daytime as discussed in Case 1. However, the PV array and battery
system was still by far the heaviest due to the large battery mass needed to operate the base camp
throughout the night. The PV array and RFC system had significantly less mass than the battery-based
system, approximately 75 percent less, but was still 30 percent heavier than the baseline reactor system
and 37 percent heavier than the scaled reactor system. As with the base camp’s power systems
comparison, for lunar missions that require high power levels throughout the nighttime period, the reactor

NASA/CR—2020-220368 77
system provides a mass advantage compared to a PV array-based system that needs to store energy for
nighttime operation.

7.0 Conclusion
The analysis presented in this report established the power requirements for operating an in situ
resource utilization (ISRU) oxygen production system and small base camp on the lunar surface
throughout 1 day and night cycle. The power requirements were then used to size a power system and
determine its mass to meet them. Three types of power systems were considered: a photovoltaic (PV)
array system using batteries for energy storage, a PV array system using a regenerative fuel cell (RFC) for
energy storage, and a 10-kWe output power-based Kilopower reactor system.
The first case examined the power system sizing for an oxygen production plant on the lunar surface.
For continuous day and night operation of the oxygen production system, the Kilopower-based reactor
system provided a significant mass advantage over the PV array-based system with either a battery or
RFC as the energy storage method. The Kilopower system’s total mass was 4,853 kg for the baseline
system providing electrical power only to the oxygen production plant. This is only 8 percent of the mass
of the PV array and battery system, which had a mass of 60,563 kg and 41 percent of the mass of the PV
array and RFC system, which had a mass of 11,789 kg. This is due mainly to the large energy storage
requirements for the PV systems needed to operate the oxygen production plant throughout the night. To
address the need for a large amount of energy storage, a variation on the production approach was
examined for the PV array-based systems. Since the goal of the oxygen production system is to produce a
certain amount of oxygen over a given time period, the production rate was increased so that the baseline
total oxygen production of 1,154 kg per 1 day and night cycle was produced only during the daytime for
the PV array-based power systems. This significantly reduced the nighttime energy storage requirement to
only “keep-alive” power required to maintain the equipment at temperatures needed to survive the lunar
night. By eliminating most of the nighttime energy storage requirement, the mass of the PV array-based
systems was significantly reduced to less than that needed for the baseline Kilopower reactor-based
system. The PV array and battery total system mass dropped to 2,740 kg and the PV array and RFC
system to 2,020 kg. These masses included the increased mass associated with scaling the ISRU oxygen
production plant to operate at the higher production rate. These masses are much less than the baseline
Kilopower system. However, the design of the oxygen production system is intended to be automated.
Therefore, it could operate without personnel in the vicinity. If that is the case, then there is the possibility
of locating the Kilopower reactor near the oxygen production system and taking advantage of the waste
heat produced by the reactor in the oxygen production process. Using this approach with the baseline
reactor system reduces the number of systems needed from three to two and reduces the mass to 3,090 kg.
This is still more than the daytime-only operation of the PV array systems, but it is more competitive. The
requirement for a two-reactor system is due to the operating temperature of the reactor and the Stirling
engine conversion efficiency. If the core operating temperature could be increased to 1,000 °C and the
Stirling engine efficiency was increased to 35 percent, this would reduce the number of reactor systems
needed to one with a system mass of 1,545 kg. This then results in a lower mass system than either the PV
array and battery or PV array and RFC systems.
The second case addressed powering a six-person base camp. The base camp power requirements
were the same for both the day and nighttime operation. Therefore, there was no means of eliminating or
reducing the energy storage requirement for the PV array-based systems. To meet the 28.05 kW
continuous power requirement, the PV array and battery system had a mass of 65,786 kg and the PV array
and RFC system had a mass of 12,805 kg. Since the camp was manned, operation of the reactor in close

NASA/CR—2020-220368 78
proximity was not possible. The only feasible reactor power system option was to operate the reactor at a
distance of 1.99 km from the base camp and provide electrical power to the base camp for all of its power
needs. Because of the distance between the reactor and the base camp, the utilization of waste heat from
the reactor would not be possible. To provide the desired power, three 10-kWe Kilopower reactor systems
would be needed with a total mass of 4,853 kg. This mass is significantly less than both of the PV array
cases. It is approximately 7 percent of the mass of the PV array and battery system and 37 percent the
mass of the PV array and RFC system.
The last case examined was a combination of powering both the ISRU oxygen production system and
the base camp. For this case, the approach considered for the PV array-based systems was producing
oxygen only during the daytime since the nighttime oxygen production powered by these systems was not
competitive masswise. The power system masses for the combined operation of the base camp and
oxygen production system for the PV array and battery and PV array and RFC systems were 64,419 and
12,731 kg, respectively. This compared to a mass of 9,909 kg for the Kilopower reactor system. To meet
the power demands of both the base camp and oxygen production, six Kilopower reactor systems were
required. However, since these are 10-kWe modular systems, this produced an excess power capacity of
6.2 kWe. To account for this, a fourth power system option was added in which the Kilopower reactor
system was scaled or optimized to match the actual power requirement of the base camp and oxygen
production. The scaled reactor system had a mass of 8,939 kg.
The results of the comparison between the solar PV array-based and Kilopower reactor-based systems
have shown that for missions required to operate throughout the lunar nighttime at power levels
comparable to those used during the daytime, the reactor-based system provides a significant mass
advantage. However, for applications that can meet their mission requirements while only having to
operate during the daytime with minimal power required to survive the nighttime, the PV array-based
systems are competitive masswise and provide a mass advantage over the current Kilopower system
design.
For the PV array-based systems, the RFC-based energy storage method had a mass advantage over a
battery-based method. For higher power nighttime power operation in all three cases, the RFC system’s
specific energy was just over 830 Wh/kg. For the lower power keep-alive nighttime power level used as
part of the analysis for the first case, the specific energy for the RFC was 456 Wh/kg. Both these levels
are significantly above the specific energy of 200 Wh/kg for the battery. Because of this higher specific
energy, the RFC-based storage system provided significant mass advantages over the battery-based
storage system.
The baseline reactor system utilized shielding and separation distance to meet the desired maximum
dose level of 5 rem/yr for personnel operating within the vicinity of the power loads, base camp, and
oxygen production system. As mentioned, in Section 5.2, there are methods that could be utilized to
reduce the shielding requirements and separation distance. Implementing these would reduce the overall
system mass for the reactor. Also, as mentioned in the third case, optimizing the reactor output for each
specific mission would provide benefits in mass at the expense of modularity.

NASA/CR—2020-220368 79
Appendix—Nomenclature
AEC Able Engineering Company, Inc.
ASRG Advanced Stirling Radioisotope Generator
Desert RATS Desert Research and Technology Studies
DOE Department of Energy
DOE–NR DOE Office of Naval Reactors
DSH Deep Space Habitat
ECLSS environmental control and life support system
HDU habitat demonstration unit
ISRU in situ resource utilization
JIMO Jupiter Icy Moons Orbiter
MLI multilayer insulation
MMW Multimegawatt Space Reactor
NEPSTP Nuclear Electric Propulsion Space Test Program
NERVA Nuclear Engine for Rocket Vehicle Applications
NNSA National Nuclear Security Administration
NREL National Renewable Energy Laboratory
PDU power distribution unit
PEM proton exchange membrane
PV photovoltaic
RAPDAR rollout and passively deployed array
RFC regenerative fuel cell
SDIO Strategic Defense Initiative Office
SNAP Systems for Nuclear Auxiliary Power
SNTP Space Nuclear Thermal Propulsion
SPAR Space Power Advanced Reactor
TRIGA Training, Research, Isotope, General Atomic
TRL technology readiness level
TSRT Thermionic System Evaluation Test
UV ultraviolet

Symbols
Aa solar array area, m2
Abe battery enclosure surface area, m2
Acp conductive path cross-sectional area, m2
Aee electrolyzer enclosure surface area, m2
Ah habitat surface area, m2
Apt passthrough area as a percent of total area
Ar radiator area, m2
Avs vibrating screen area, m2
als lunar surface albedo

NASA/CR—2020-220368 81
Bdod battery depth of discharge
Bw communications transmission bandwidth, Hz
Ca radiator adhesive scaling coefficient, kg/m2
Cas insulation attachment and seals scaling coefficient, kg/m2
Cat radiator attachment scaling coefficient, kg/m2
Cc radiator coating scaling coefficient, kg/m2
Ccf cold plate coverage coefficient
Ch radiator header scaling coefficient, kg/m2
Chp heat pipe mass scaling coefficient, kg/m
Cic insulation inner cover scaling coefficient, kg/m2
Clv radiator louvers scaling coefficient, kg/m2
Coc insulation outer cover scaling coefficient, kg/m2
Cp radiator panel scaling coefficient, kg/m2
Crl insulation reflective layer scaling coefficient, kg/m2
Cs radiator stingers scaling coefficient, kg/m2
Csp insulation spacer scaling coefficient, kg/m2
Ct radiator tubing scaling coefficient, kg/m2
c speed of light, m/s
cp specific heat of ilmenite, (J/kg)K
dh habitat diameter, m
dhr distance to the horizon, m
dH2t hydrogen tank diameter, m
di regolith particle size diameter, m
dicp conductive path inner diameter, m
dil spacing between insulation layers, m
dn day number, Earth days
dnt total number of Earth days in a lunar year
dO2t oxygen tank diameter, m
dr oxygen production reactor diameter, m
drs travel distance of scoop in regolith, m
ds maximum regolith particle diameter, m
dt tank diameter, m
dtr communications transmitter to receiver distance, m
dtrmax maximum transmission distance, m
d1 radiation dose at position 1, rem/yr
d2 radiation dose at position 2, rem/yr
Ea energy produced by the solar array, Wh
Ed daytime energy requirement, Wh
En nighttime energy requirement, Wh
Fal radiator louver area adjustment factor

NASA/CR—2020-220368 82
fcl centerline frequency, Hz
fls view factor to the lunar surface
fn passthrough constant 1
fp passthrough constant 2
fs material yield strength factor of safety
fsc solar cell fill factor
Gw wire gauge
g lunar surface gravity, m/s2
gr ratio of lunar to Earth gravitational constants
Hr oxygen production heat of reaction, kJ/kg
h height above the surface, m
hr oxygen production reactor height, m
hre communications receiver height, m
hrs height of the oxygen production reactor above the surface, m
ht communications transmitter height, m
Im mean solar intensity at the Earth/Moon system, W/m2
Is solar intensity at the Earth/Moon system, W/m2
Iw wire current, A
K transmission adjustment factor
k thermal conductivity o material, W/(m·K)
kb Boltzmann constant, J/K
Lc cold plate length, m
Lcb regolith and ilmenite conveyor belt length, m
Lcp conductive path length, m
Lw length of the wire runs, m
Mab solar array and battery system total mass, kg
Mapv solar array blanket mass, kg
Marfc solar array and RFC system total mass, kg
Mas solar array structure mass, kg
Mb battery mass, kg
Mbb breaker box mass, kg
Mbcc battery charge controller mass, kg
Mcb conveyor belt mass, kg
Mcp cold plate mass, kg
Me electrolyzer mass, kg
Mes energy storage mass, kg
Mfc fuel cell mass, kg
Mfr fission reactor mass, kg
Mfrb fission reactor balance-of-plant mass, kg
Mfrs fission reactor shield mass, kg

NASA/CR—2020-220368 83
Mfrt fission reactor system total mass, kg
MH2 hydrogen mass, kg
MH2t hydrogen tank mass, kg
Mhp heat pipe mass, kg
Mmli insulation mass, kg
Mms magnetic separator mass, kg
M O2 oxygen mass, kg
MO2t oxygen tank mass, kg
Mr radiator mass, kg
Mrfc regenerative fuel cell mass, kg
Mrt oxygen production total reactor tank mass, kg
Msw specific mass of the wire, kg/m
Mt reactor tank mass, kg
Mtc thermal control system mass, kg
Mvc direct current (dc) to dc voltage converter mass, kg
Mvs vibrating screen mass, kg
Mw wire mass, kg
M i rate at which ilmenite is separated from the regolith, kg/h
M o rate of oxygen production, kg/h
M r rate at which regolith is acquired from the surface, kg/h
M w rate of water production, kg/h
ncp number of conductive paths
nl number of layers of insulation
nri regolith particle size distribution
nrs number of fission reactor systems
nrw number of wires per fission reactor system
nw number of wire runs between array and junction box
Pa solar array output power, W
Pahc habitat air conditioning power, W
Pamax maximum solar array output power, W
Pbc base camp power requirement, W
Pbm biomass processing power, W
Pcom communications power requirement, W
Pdl daytime power level required by the loads, W
Pe electrolyzer power, W
Peva extravehicular activity support power, W
Pfd food processing and storage power, W
PH2 hydrogen gas pressure, Pa
Pht habitat nighttime heater power, W
Pil power to lift the ilmenite, W

NASA/CR—2020-220368 84
Pls environmental control and life support power, W
Pms magnetic separator power, W
Pn noise power level, W
Pnl nighttime power level required by the loads, W
Poe operational equipment power, W
Pr heat rejection power to the radiator, W
Prl power to lift regolith, W
Pro surface-to-satellite receiver power, W
Prs power to scoop regolith, W
Pss communications signal strength, W
Pt tank internal pressure, Pa
Pte electrolyzer waste heat, W
Pth habitat thermal control power, W
Ptm communications transmitter power, W
PtO2 total oxygen production system electrical power required, W
Pvs vibrating screen power, W
Pwl transmission wire power loss, W
Pwp water processing power, W
Pwst waste processing power, W
Qas airlock structure heat leak to the surroundings, W
Qcp conductive path heat loss, W
Qh habitat total nighttime heat loss, W
Qhi thermal power required to heat the ilmenite, W
Qi insulation heat loss, W
Qls oxygen production reactor heat loss to the surroundings, W
Qps passthrough heat loss, W
Qr heat consumed by the oxygen production process, W
Qra heat leak to the radiator during nighttime, W
Qsl support leg heat leak to the surface, W
QtO2 total oxygen production system thermal power required, W
RH2 hydrogen gas constant
Rm ratio of mass of ilmenite to mass of regolith
Rmax maximum achievable communications data rate, bit/s
Rm′ average amount of small grain ilmenite in mare by weight
Rres hydrogen and oxygen reactant residual fraction
Rv′ average amount of small grain ilmenite in mare by volume
Rw power transmission wire resistance per length, Ω/m
Rvs regolith screening rate, kg/(h·m2)
rm radius of the Moon, km
rorb solar orbital radius of the Earth/Moon system, km

NASA/CR—2020-220368 85
rorbm average solar orbital radius of the Earth/Moon system, km
r1 distance from reactor for dose level 1, m
r2 distance from reactor for dose level 2, m
Sa spacing between the array modules, m
Sb battery specific energy, Wh/kg
Schv check valve specific mass, kg/kW
Scu controller unit specific mass, kg/kW
Scv control valve specific mass, kg/kW
Ses electrolyzer stack specific mass, kg/kW
Sf filter specific mass, kg/kW
Sfc fuel cell stack specific mass, kg/kW
Sfr flow regulator specific mass, kg/kW
Sfs flow sensor specific mass, kg/kW
She heat exchanger specific mass, kg/kW
Shps hydrogen phase separator specific mass, kg/kW
Shr hydrogen regulator specific mass, kg/kW
Shst fuel cell hydrogen separation tank specific mass, kg/kW
Sor oxygen regulator specific mass, kg/kW
Sost oxygen separation tank specific mass, kg/kW
Spl propellant line and fittings specific mass, kg/kW
Srti oxygen production reactor tank scaling factor
Ss pressure and temperature sensor specific mass, kg/kW
Svb battery specific volume, kg/m3
Sve electrolyzer specific volume, kg/m3
Sw wiring specific mass, kg/kW
Swp water pump specific mass, kg/kW
Swt electrolyzer water tank specific mass, kg/kW
TH2O electrolyzer water temperature, K
Thi habitat internal temperature, K
Ti reactor ilmenite and hydrogen temperature, K
Tio regolith surface temperature at collection time, K
T1s maximum lunar surface temperature, K
Tm mean insulation temperature, K
Tr radiator operating temperature, K
Ts sink temperature, K
Tsd lunar surface average daytime sink temperature, K
Tsn lunar surface average nighttime sink temperature, K
tc cold plate thickness, m
tcp conductive path wall thickness, m
td lunar day length, h

NASA/CR—2020-220368 86
th time to heat ilmenite, h
ti instantaneous day time, h
trt time ilmenite is maintained at the reaction temperature, h
tsc fraction of batch processing time used to scoop the regolith
tt tank wall thickness, m
UH2 hydrogen gas velocity within the oxygen production reactor, m/s
Va array operating voltage, V
VH2t hydrogen tank volume, m3
VO2t oxygen tank volume, m3
Vrt power wire voltage, V
Vtm tank material volume, m3
Vir rate ilmenite is replaced in the oxygen production reactor, kg/hr
wc cold plate width, m
ZH2 hydrogen gas compressibility factor
α solar elevation angle
αrls radiator lunar infrared (IR) absorptivity
αrs radiator solar infrared (IR) absorptivity
β array inclination angle
γ day angle
∆G Gibbs free energy, kJ/mol
∆H change in enthalpy, kJ/mol
∆S change in entropy, (kJ/K)mol
∆SH2 hydrogen entropy change, (J/K)mol
∆SH2O water entropy change, (J/K)mol
∆SO2 oxygen entropy change, (J/K)mol
ε eccentricity of the Earth/Moon system
εhw habitat wall emissivity
εi insulation layer emissivity
εl lunar orbital eccentricity
εp fluidized bed porosity
εr radiator emissivity
εrO2 oxygen production reactor emissivity
ηbc battery charge efficiency
ηc energy storage system charge efficiency
ηd energy storage system discharge efficiency
ηe electrolyzer efficiency
ηem electric motor efficiency
ηfc fuel cell efficiency
ηs ilmenite separation process efficiency
ηsc solar cell efficiency

NASA/CR—2020-220368 87
θ hour angle
θrs maximum solar incident angle to the radiator
µH2 hydrogen gas viscosity, kg/(m·s)
µr regolith coefficient of friction
ρa solar array areal density, kg/m2
ρas solar array structure specific mass, kg/m2
ρc cold plate material density, kg/m3
ρH2 hydrogen gas density, kg/m3
ρi granular ilmenite density, kg/m3
ρip ilmenite particle density, kg/m3
ρr average regolith density, kg/m3
ρtm tank material density, kg/m3
σ Stefan-Boltzmann constant, W/(m2·K4)
σy tank material yield strength, Pa
τ ilmenite particle shape factor
φ latitude
χd percent dust coverage
ψ lunar declination angle in degrees
ψmax maximum lunar declination angle in degrees

NASA/CR—2020-220368 88
References
1. National Aeronautics and Space Administration: Explore Moon to Mars: Earth’s Moon. 2009.
https://www.nasa.gov/multimedia/imagegallery/image_feature_1538.html Accessed Nov. 4, 2019.
2. Bussey, D.B.J., et al.: Permanent Sunlight at the Lunar North Pole. Presented at the Lunar and
Planetary Science XXXV Meeting, Houston, TX, 2004.
3. Choi, Charles Q.: Moon Facts: Fun Information About the Earth's Moon. Space.com, 2017.
https://www.space.com/55-earths-moon-formation-composition-and-orbit.html Accessed Nov. 7,
2019.
4. Berry, Patrick L.: A Hithchiker’s Guide to the Moon. 2005.
https://www.nasa.gov/vision/universe/solarsystem/24may_lola.html Accessed Nov. 1, 2019.
5. Jet Propulsion Laboratory: Solar Cell Array Design Handbook: Volume 1. JPL SP 43–38, 1976.
6. Williams, J.-P., et al.: The Global Surface Temperatures of the Moon as Measured by the Diviner
Lunar Radiometer Experiment. Icarus, vol. 283, 2017, pp. 300‒325.
7. Smith, R.E.; and West, G.S.: Space and Planetary Environment Criteria Guidelines for Use in Space
Vehicle Development, 1982 Revision (Volume 1). NASA TM‒82478, 1983. http://ntrs.nasa.gov
8. Eagle Engineering, Inc.: Conceptual Design of a Lunar Oxygen Pilot Plant Lunar Base Systems Study
(LBSS) Task 4.2. EEI Report 88–182, 1988.
9. Colozza, Anthony J.: Analysis of Lunar Regolith Thermal Energy Storage. NASA CR–189073, 1991.
http://ntrs.nasa.gov
10. Colozza, Anthony J.: Hydrogen Storage for Aircraft Applications Overview. NASA/CR—2002-
211867, 2002. http://ntrs.nasa.gov
11. Weiss, Norman L., ed.: SME Mineral Processing Handbook. Vols. 1 and 2, American Institute of
Mining, Metallurgical, and Petroleum Engineers, Englewood, CO, 1985.
12. Guangzai, Nong; Yijing, Li; and Yin, Yongjun: Energy Analysis on the Water Cycle Consisting of
Photo Catalyzing Water Splitting and Hydrogen Reacting With Oxygen in a Hydrogen Fuel Cell.
Chem. Phys. Lett. X, vol. 4, 2019.
13. Hanford, Anthony J.: Advanced Life Support Research and Technology Development Metric—Fiscal
Year 2005. NASA/CR—2006-213694, 2006. http://ntrs.nasa.gov
14. Bhasin, Kul; and Hayden, Jeffrey L.: Evolutionary Space Communications Architectures for
Human/Robotic Exploration and Science Missions. NASA/TM—2004-213074, 2004.
http://ntrs.nasa.gov
15. Lindsey III, Jefferson F.: Lunar Surface Transmission Loss for the Apollo Astronaut. NASA TN D‒
4915, 1968. http://ntrs.nasa.gov
16. Stallings, William: Data and Computer Communications. Macmillan Publishing Co., New York, NY,
1985.
17. National Renewable Energy Laboratory: Photovoltaic Research. 2019. https://www.nrel.gov/pv/
Accessed Jan. 8, 2019.
18. Barrett, Rory, et al.: Development of a Passively Deployed Roll-Out Solar Array. AIAA 2006‒4011,
2006.
19. Duchek, Matthew E., et al.: Self-Deploying Tent Array for Mars Surface Solar Power. AIAA 2018‒
1943, 2018.
20. Crutchik, M.; Colozza, Anthony J.; and Applebaum, J.: A Photovoltaic Catenary-Tent Array for the
Martian Surface. NASA TM‒106306, 1993. http://ntrs.nasa.gov
21. Epec Engineered Technologies: Lithium Battery Technologies and Battery Chemistry. 2019.
https://www.epectec.com/batteries/ Accessed Nov. 4, 2019.

NASA/CR—2020-220368 89
22. Luo, Xing, et al.: Overview of Current Development in Electrical Energy Storage Technologies and
the Application Potential in Power System Operation. Appl. Energy, vol. 137, 2015, pp. 511‒536.
23. Mallon, Kevin R.; Assadian, Francis; and Fu, Bo: Analysis of On-Board Photovoltaics for a Battery
Electric Bus and Their Impact on Battery Lifespan. Energies, vol. 10, 2017, p. 943.
24. Herron, David: Gasoline, Electricity, and the Energy to Move Transportation Systems, Green
Transportation. 2019. https://greentransportation.info/energy-transportation/introduction.html
Accessed Feb. 1, 2019.
25. Idaho National Laboratory: Atomic Power In Space II. INL/EXT‒15‒34409, 2015.
26. NASA Space Technology Mission Directorate: Kilopower. 2019.
https://www.nasa.gov/directorates/spacetech/kilopower Accessed Nov. 4, 2019.
27. NASA Facts: Space Technology: Game Changing Development: The Fission System Gateway to
Abundant Power for Exploration. FS‒2018‒01‒291‒LaRC, 2018.
https://www.nasa.gov/sites/default/files/atoms/files/ns_kilopower_fs_180111.pdf Accessed Nov. 4,
2019.
28. NASA COMPASS Team: ISRU Power System Demonstrator: Final Draft. 2016.
29. United States Nuclear Regulatory Commission: Information for Radiation Workers. 2018.
https://www.nrc.gov/about-nrc/radiation/health-effects/info.html Accessed Nov. 4, 2019.

NASA/CR—2020-220368 90