The 1980s Lunar Revival: Lunar Oxygen (1983)

Climbing toward reusability: liftoff of the Space Shuttle Orbiter Columbia at the start of mission STS-2 (12-15 November 1981). Image credit: NASA.

At the heart of the Space Transportation System (STS) was the Space Shuttle. The first Space Shuttle mission, STS-1 (12-14 April 1981), was the first two-person Orbiter Flight Test (OFT) mission and the first flight of the Shuttle Orbiter Columbia. The second OFT, STS-2 (12-15 November 1981), had to be cut to two days in orbit from a planned five following the failure of one of Columbia's three electricity-producing fuel cells. Nevertheless, STS-2, the first reflight of a reusable spacecraft and the first flight of the Canada-built Remote Manipulator System (RMS) robot arm, was viewed as a success.

When Columbia glided to a landing for the second time, the form the STS would eventually take was still poorly defined. It would remain so at least until the destruction of the Shuttle Orbiter Challenger (28 January 1986) at the start of the STS-51L, the 25th flight of the Shuttle Program. The loss of Challenger and her seven-member crew marked the end of the optimistic first phase of the Space Shuttle Program.

Before that, however, the STS seemed ripe for augmentation. It would, of course, include expendable rocket stages attached to satellites carried to low-Earth orbit (LEO) in the Shuttle Orbiter Payload Bay; these "upper stages" were meant mainly to boost payloads from LEO to geosynchronous orbit (GEO), but could also launch robotic spacecraft from LEO on interplanetary trajectories. In addition, the STS would include Spacelab, a European-built system of Payload Bay-mounted laboratory modules and scientific instrument pallets. Development of upper stages and Spacelab had commenced in the 1970s, shortly after Space Shuttle development began.

Many saw the stages and Spacelab as interim steps toward more complex and competent STS elements. The former, it was expected, would lead to a reusable Orbital Transfer Vehicle (OTV); the latter, to a Space Station assembled in LEO from Orbiter-launched components. The OTV and its "little brother," the Orbital Maneuvering Vehicle (OMV), were typically seen as auxiliary vehicles based at the Space Station.

NASA Johnson Space Center (JSC) in Houston, Texas, anticipated that the Space Station would support ambitious space construction projects; for example, large communications platforms in GEO. In 1979, inspired in part by its involvement in joint Department of Energy/NASA Solar Power Satellite studies, JSC studied a Space Station concept it called the Space Operations Center (SOC). After an initial flurry of planning activity, Shuttle delays put the SOC on the back burner; immediately after STS-1, however, JSC efforts to promote the assembly base in LEO kicked into high gear.

The Space Operation Center (SOC), a Shuttle-launched low-Earth orbit (LEO) space station, as it was envisioned in 1982. Intended as an assembly and repair base, the SOC would have included a "surrogate" Shuttle Orbiter Payload Bay (lower center) for satellite servicing and a hexagonal hangar (center right) for storing and servicing the Orbital Transfer Vehicle (OTV) and Orbital Maneuvering Vehicle (OMV). Image credit: NASA. 

A remote-controlled Orbital Maneuvering Vehicle (OMV), a small "space tug," closes in on the Gamma Ray Observatory (GRO) in this illustration from 1986. After its launch in 1991, GRO was renamed the Compton GRO. The reusable modular OMV, intended as an auxiliary vehicle for extending Shuttle Orbiter and Space Station capabilities, was cancelled in 1987 in the aftermath of the Challenger accident. Image credit: TRW/NASA.

JSC is widely known as the home base of the astronauts and site of the Mission Control Center. Less well known, perhaps, is its long-time contribution to lunar and planetary science. The Lunar Receiving Laboratory (LRL), built in 1967, was used for analysis and storage of lunar geologic samples beginning with Apollo 11 in July 1969. JSC also played a key role in organizing the annual Lunar Science Conference (LSC), the first of which was held in Houston in January 1970.

At the time, JSC was called the Manned Spacecraft Center (MSC). It was renamed in 1973 after the death of President Lyndon Baines Johnson. The LSC was renamed the Lunar and Planetary Science Conference (LPSC) in 1978.

Michael Duke was on hand when the Apollo 11 samples arrived at the LRL; he was Lunar Sample Principal Investigator for the mission. In July 1969, he had been working for the U.S. Geological Survey (USGS) Branch of Astrogeology for six years. In 1970, he left USGS Astro (as ambitious people are wont to do) to become Lunar Sample Curator at NASA MSC, a post he held until 1977, when he became Chief of the NASA JSC Planetary and Earth Sciences Division.

Shortly after STS-2, Duke and another scientist in his division, Wendell Mendell, became concerned that developing the SOC and other proposed STS elements might mean reduced funding for NASA science programs. Space science at NASA was already hurting; the new Administration of President Ronald Reagan had made deep cuts. Rather than oppose new STS elements, Duke and Mendell sought ways that the SOC, OTV, and other proposed hardware could advance scientific exploration. Specifically, they sought to make the case for a base on the Moon.

Their efforts in some ways paralleled those of lunar scientists at the dawn of the Apollo Program, when lunar science barely existed as a field of study. Much like those early pioneers, Duke and Mendell sought to find and bring together individuals and organizations to build a constituency. Initially, they found prospective lunar base allies through informal, low-profile contacts. By late 1982, however, it was time to go public.

This they did by organizing three public special sessions at the 14th LPSC, which was held at NASA JSC in March 1983. The lunar sessions were titled "Return to the Moon" and "Future Lunar Programs." The third session, "Prospects for Planetary Exploration," sought to tie their lunar base efforts to the interests of the broader Solar System exploration community.

In their introduction to the lunar special sessions abstract volume, Duke and Mendell explained that "very little vision is required to see the [STS] reaching to the Moon." They argued that "the lunar option requires decisions today — but not dramatic ones." They pitched a Fiscal Year (FY) 1985 start for the lunar base program, but took pains to stress that the lunar base would need little or no new dedicated funding before FY 1991 or FY 1992.

In fact, they expected that lunar capability would grow more or less naturally from the STS in the late 1990s, several years after the SOC, OTV, and OMV were in place to support GEO missions. The amount of energy required to put a satellite into GEO is, after all, very nearly the same as that needed to put a payload into low-lunar orbit (LLO).

The lunar special sessions abstract volume included 22 abstracts by more than 30 authors and co-authors. The abstracts covered topics ranging from philosophy, law, and economics to geology, physiology, and energy. Of particular interest was an abstract by Hubert Davis, Senior Vice President of Houston-based Eagle Engineering, Incorporated (EEI). In it, he proposed to extend the STS to the Moon. His aim: to make the STS more economical by mining, refining, and using as rocket propellant oxygen chemically locked in lunar dirt.

EEI had been established by retired NASA JSC engineers. Davis, a co-founder of the company, began his career as an aircraft maintenance engineer in the U.S. Air Force during the Korean War. Inspired by President John F. Kennedy's May 1961 call for a man on the Moon, he went to work for the MSC Power & Propulsion Division in March 1962. He oversaw the test program for Lunar Module 5; better known as Eagle, it bore Apollo 11 astronauts Neil Armstrong and Edwin "Buzz" Aldrin to the lunar surface on 20 July 1969.

As Apollo ended, Davis transferred to the MSC Special Projects Office, where he studied new STS elements — cheap solid-propellant STS upper stages and a Shuttle-derived heavy-lift launcher — as well as Solar Power Satellites manufactured from lunar materials. He took early retirement in 1978 after being made the JSC Engineering Directorate representative to the Space Shuttle Program Office; in a 2009 NASA JSC oral history interview, Davis explained that he had left the space agency at age 48 because he found the Shuttle oversight job to be boring.

The industry magazine Aviation Week & Space Technology made his presentation the centerpiece of its coverage of the LPSC lunar base special sessions. Two months following the special sessions Davis published an expanded version of his abstract and presentation as an EEI report called Lunar Oxygen Impact Upon STS Effectiveness.

Davis acknowledged that his study was incomplete and preliminary. He did not, for example, examine how his oxygen mining and refining facility would be established on the Moon. That the study was preliminary explained why it was incomplete; if it could not provide an early indication that lunar oxygen would enhance STS capabilities, Davis argued, then there would be little point in conducting a more detailed study of the concept.

To illustrate how lunar-produced liquid oxygen might enhance the STS, Davis used as an example STS-40, which was scheduled to place the Galileo spacecraft into LEO on 30 May 1986. Galileo would reach LEO attached to an expendable Centaur G' upper stage which, following release from the Orbiter payload bay, would launch it out of LEO on a direct trajectory to Jupiter.

The Galileo spacecraft was expected to weigh 2510 kilograms and the Centaur G' without propellants, 2650 kilograms. Support equipment for maintaining Galileo and Centaur G' in the payload bay would weigh 470 kilograms and 3640 kilograms, respectively. Filling the Centaur G' large tank with low-density liquid hydrogen fuel would add 3310 kilograms to STS-40's payload weight; filling the Centaur G' small tank with dense liquid oxygen oxidizer would increase payload weight by a whopping 16,570 kilograms. STS-40 payload weight thus came to 29,480 kilograms, with liquid oxygen making up 56% of the total.

The Galileo Jupiter spacecraft and an expendable Centaur G' upper stage move away from the Shuttle Orbiter that deployed them in low-Earth orbit. The large-diameter forward section of the stage (center), to which Galileo is attached, contains low-density liquid hydrogen fuel; the small-diameter aft section, to which are attached two rocket motors, high-density liquid oxygen oxidizer. Image credit: NASA.

Davis then imagined a 2510-kilogram Shuttle-launched payload in the first decade of the 21st century, after the STS had been extended to the Moon. A reusable OTV would replace the Centaur G'. Permanently basing the OTV at the SOC meant that it would not add to Shuttle payload weight.

The Shuttle Orbiter would carry a 660-kilogram tank containing 3310 kilograms of liquid hydrogen for fueling the OTV in LEO. The propellant dump in LEO near the SOC would provide the OTV with lunar liquid oxygen. Support equipment in the payload bay for the spacecraft and hydrogen tank would bring the total Shuttle payload weight to just 7280 kilograms, or about one quarter of the STS-40 total.

A chemical process called hydrogen reduction of ilmenite formed the basis of the lunar oxygen STS infrastructure. Ilmenite (chemical formula FeTiO3), a titanium ore, is a mineral common in the basaltic rocks, dirt, and dust that form the dark-hued lunar plains known as maria (Latin for "seas").

Davis focused on ilmenite rather than lunar polar ice — which can provide both hydrogen and oxygen — because in 1983 no one knew that water ice exists at the lunar poles. Though the lunar polar ice hypothesis was by then more than 20 years old, the first evidence that it might be correct would not be found until 1994, when Clementine became the first spacecraft to explore the Moon from lunar polar orbit.

Mining robots would continuously gather ilmenite-rich lunar dirt at a rate of 28 metric tons per hour and deliver it to a separation facility. The dirt would first be sieved to remove large dirt particles, clods, and rocks. The resulting fine-grained dirt and dust would then be heated and subjected to an electrostatic process that would separate the ilmenite at a rate of 2.27 metric tons per hour.

The ilmenite would be moved to the hydrogen-reduction unit, where it would be exposed to hydrogen gas at 700° Celsius (C) and 2.7 Earth atmospheres of pressure. The hydrogen would bind with and free the oxygen bonded to the iron. This would yield water vapor at a rate of 0.26 metric tons per hour, which would be cooled, condensed, and subjected to electrolysis, splitting it into oxygen and hydrogen.

The oxygen would be chilled until it condensed into dense liquid, then stored in spherical tanks. The hydrogen would be returned to the reduction unit for reuse and the powdery titanium oxide and iron left over from the reduction process — about 90% of the original mass of the lunar dirt — would be stacked out of the way for possible future use.

The facility would use a little more than six megawatts of electricity continuously; this might be reduced if waste heat from the refining process could be exploited effectively. Davis estimated that his mining and refining facility could produce 150 metric tons of liquid oxygen per month.

The Aft Cargo Carrier (ACC) — the blue short cylinder and truncated cone attached at left to the bottom of the orange Space Shuttle External Tank (ET) — would augment the 4.57-by-18.28-meter Shuttle Orbiter Payload Bay (the blue cylinder within the Orbiter outlined by dashed lines). The 9.72-meter-long, 8.38-meter-wide ACC would, among other things, facilitate launch of OTV components and spherical tanks filled with liquid hydrogen. Image credit: Martin Marietta.

This painting by Eagle Engineering artist Pat Rawlings displays elements of Davis's proposed lunar oxygen STS infrastructure in LEO. The close proximity of the elements is schematic, not realistic. At lower right, a remote-controlled OMV detaches a spherical tank filled with liquid hydrogen from an ET/ACC. The LEO propellant dump is at lower left. Small in the distance above it, silhouetted against the Earth, is a Shuttle Orbiter. A version of the SOC is depicted to the right of the Orbiter. In the foreground, a small OTV prepares to leave Earth orbit; liquid hydrogen tanks bound for lunar orbit ride on its bowl-shaped rigid aerobrake heat shield.

Davis described his lunar oxygen STS infrastructure in operation. A Space Shuttle would launch liquid hydrogen for the LEO propellant dump in a spherical tank inside an Aft Cargo Carrier (ACC) fitted over the dome-shaped end of its External Tank (ET).

Normally, the ET would separate from the Orbiter as it neared orbital velocity, tumble and break up in the upper atmosphere, and fall into the Indian Ocean. When the ACC was attached, however, the Shuttle Orbiter would boost the ET/ACC combination to LEO and separate. A remote-controlled OMV based at the SOC would then detach the hydrogen tank from the ACC and move it to the propellant dump, where refrigeration and high-tech insulation would ensure that no hydrogen was lost to boil-off.

Zero liquid hydrogen boil-off was, Davis explained, critical to making his lunar oxygen STS infrastructure viable. He wrote that the Centaur G' stage was expected to lose about 3% of its liquid hydrogen to boil-off per day. A similar boil-off rate at any point in his lunar oxygen STS infrastructure would be "intolerable."

Davis assumed two types of modular OTVs, each of which could be tailored to carry out several types of missions. The OTVs, clusters of spherical propellant tanks linked by struts, would perform roundtrip missions between LEO and GEO and between LEO and an LLO SOC and propellant dump. The OTVs could operate with or without a pressurized module containing a crew.

The smaller OTV, which would burn 25 metric tons of liquid hydrogen and liquid oxygen during a voyage from LEO to LLO and back, would include a rigid aerobrake heat shield 18 meters wide. The heat shield, which would include thermal protection tiles akin to those attached to the Space Shuttle Orbiter, would enable the OTV to use atmospheric drag to capture into LEO with minimal propellant expenditure. The smaller OTV could transport nearly 43 metric tons of liquid hydrogen from the LEO propellant dump to its counterpart in LLO.

A single-stage Lunar Module lander based on the smaller OTV design would burn 28 metric tons of liquid hydrogen/liquid oxygen propellants to travel from LLO to the lunar surface and back. It would be capable of lifting 41 metric tons of liquid oxygen from the lunar surface to the LLO propellant dump.

Davis used Lunar Module landing gear as an example of how hardware in his lunar oxygen STS infrastructure would need to be optimized to reduce mass. The Apollo Lunar Module's four landing legs and foot pads accounted for 3.3% of its landed weight; the small OTV-based Lunar Module would exploit new materials and improved understanding of the lunar surface to reduce the figure to 2%.

On the Moon: at lower left, a robotic front-end loader scoops lunar dirt; behind it, another deposits dirt in a hopper at the start of the lunar oxygen refining process. Cables strung on poles link the lunar oxygen refinery to a nuclear reactor just over the horizon. Lunar oxygen is liquified and poured into a tank at center right; the filled tank will be added to the stack of tanks in the background at center top. Conveyor belts transport tailings to a storage area at upper left, just beyond the Lunar Module launch & landing pad, and to the open pit mine at center left. Image credit: Pat Rawlings/Eagle Engineering, Incorporated/NASA.

In lunar orbit: in the foreground, a large OTV with a stowed white ballute heat shield prepares to depart LLO for Earth orbit carrying a cargo of lunar oxygen. In the background, the LLO propellant dump orbits close by the lunar SOC. Meanwhile, a Lunar Module bearing lunar liquid oxygen moves in to dock with the propellant dump. Image credit: Pat Rawlings/Eagle Engineering, Incorporated/NASA.

In its basic form, the larger OTV would carry a propellant load of 33 metric tons. Two such OTVs could be combined to form an OTV with a propellant load of 78 tons. The latter configuration would be capable of transporting more than 200 metric tons of lunar oxygen from LLO to the LEO propellant dump. This would, he calculated, require an aerobrake heat shield about 115 meters wide; that is, wider than an American football field with end zones.

One might be forgiven for asking why such a large lunar oxygen cargo was necessary; that is, why Davis did not propose transporting it to LEO using several smaller OTVs spaced out over time. He explained that minimum-energy opportunities for travel from the LLO propellant dump to LEO would occur less than once per month. They would be infrequent because the OTV could only depart LLO as its orbital plane coincided with that of the LEO propellant dump. To do otherwise would demand plane-change maneuvers that would contribute toward making the lunar oxygen STS infrastructure uneconomical.

During aerobraking, the OTV would pass through Earth's upper atmosphere at an altitude of between 50 and 100 kilometers so that atmospheric drag could reduce its speed. The OTV would then climb back into space toward an apogee (orbit high point) near SOC altitude (about 400 kilometers). At apogee, it would ignite its engines to raise its perigee (orbit low point) out of Earth's atmosphere. For the perigee-raise maneuver, Davis budgeted only enough propellants to change OTV speed by 100 meters per second. He suggested that, if further analysis showed this to be insufficient, then an SOC-based OMV might retrieve the OTV and lunar oxygen payload at apogee.

In neither the small OTV nor the large OTV case could aerobrake heat shield mass exceed 3.5% of OTV mass at the time of Earth atmosphere entry. Davis focused on heat shield mass reduction because other OTV systems were already optimized, OTV propellants had been trimmed to the bare minimum required, and reducing the liquid oxygen cargo would defeat the purpose of the exercise. He conceded that cutting aerobrake heat shield mass so dramatically might constitute a significant technical challenge; most OTV studies, he explained, had assumed that the heat shield would make up at least 10% of OTV mass at Earth atmosphere entry.

Ballute in action. This is representative of ballutes in general; the inflatable heat shield is shown here attached to a single-engine cylindrical tug, not the two-engine large OTV Davis described, and aerobraking events take place at higher altitudes than in his scenario. Image credit: NASA.

To reduce the mass of the large OTV heat shield, Davis suggested that it might take the form of an expendable fabric ballute ("balloon-parachute"). The OTV would point its twin engines in its direction of motion as the donut-shaped ballute inflated; the engines would then operate in "idle" mode to create a relatively cool gas barrier between the ballute surface and the high-temperature plasma generated in front of the ballute by Earth atmosphere reentry at lunar-return speed (3.2 kilometers per second).

Davis used computer models to attempt to determine the Mass Payback Ratio (MPR) of his proposed lunar oxygen STS infrastructure. An MPR of 1 would mean that the mass of resources (mainly propellants) expended to exploit lunar oxygen would equal the mass of the lunar oxygen supplied to LEO. NASA would thus gain nothing from putting lunar oxygen to work in the STS. If, on the other hand, the mass of the lunar oxygen delivered to LEO exceeded the mass of the resources needed to exploit it, then more detailed study might be justified.

Davis cited a computer model that included 25 roundtrip OTV flights between LEO and LLO and 103 roundtrip Lunar Module flights between LLO and the lunar surface. He wrote that, in exchange for 983 metric tons of liquid hydrogen, hydrogen tanks, and OTV attitude-control system propellant dispatched to the Moon, 2414 metric tons of lunar liquid oxygen would arrive in the LEO. He judged that this quantity could support 90 OTV flights between LEO and GEO over a period of about five years.

This indicated a preliminary MPR of 2.45, which, Davis wrote, justified additional study. He anticipated, however, that it probably would not provide enough margin to maintain a positive MPR if the mass of hardware and propellants required to establish and maintain the lunar oxygen STS infrastructure were taken into consideration.

Davis did not provide weight estimates for the LEO propellant dump, the LLO propellant dump and LLO SOC, and the Lunar Modules. Neither did he estimate the weight of the Earth-launched liquid hydrogen and liquid oxygen propellants needed to initiate the lunar oxygen STS infrastructure, nor the weight of Earth-launched liquid hydrogen needed to fly resupply and crew rotation missions after lunar oxygen became available. He assumed that the OTVs and LEO SOC would be built for LEO and GEO operations even if NASA did not return to the Moon, so disregarded their weight in his model.

He did, however, provide a weight estimate for the lunar surface mining and refining facility. Mining robots, a habitat for housing 10 facility caretakers, refining equipment, storage tanks, a nuclear reactor for generating electricity, radiator panels, and other equipment would have a combined weight of 437 metric tons. Adding this to the 983 tons of hydrogen, tanks, and attitude-control propellant would lead to an MPR of only 1.7.

If, somehow, the MPR remained sufficiently favorable after more detailed technical studies yielded credible complete weight estimates, then complex economic analyses would follow. These would, Davis explained, be based on real-world dollars and would take into account societal factors such as "affordability."

Davis conceded that extending the STS to the Moon probably could not be justified solely on the basis of economics. He argued that lunar resources should nevertheless be developed. He cited a January 1982 Los Alamos National Laboratory (LANL) proposal for an international research laboratory on the Moon; it promised wide-ranging scientific, economic, political, and defense benefits. With a nod to the political language of the 1980s United States, Davis declared that the "vitality of Free World commerce and physical security would be greatly increased by the presence of. . .resources in space."

Sources

Space Operations Center presentation materials, NASA Johnson Space Center, 18 January 1982.

"NASA Conference to Highlight Return to the Moon," NASA News Release 83-007, Steve Nesbitt, no date (March 1983).

"Economic Benefits of Lunar Base Cited," E. Bulban, Aviation Week & Space Technology, 18 April 1983, pp. 132-133, 135-137.

Fourteenth Lunar and Planetary Science Conference Special Sessions Abstracts — Return to the Moon — March 16, 1983, Future Lunar Program — March 17, 1983, LPI Contribution 500, Lunar and Planetary Institute, 1983.

Lunar Oxygen Impact Upon STS Effectiveness, Report No. 8363, Hubert Davis, Eagle Engineering, Inc., May 1983.

"Return to the Moon," Andrew Chaikin, Sky & Telescope, June 1983, p. 493.

NASA Johnson Space Center Oral History Project Edited Oral History Transcript: Hubert P. Davis, 28 July 2009 (https://historycollection.jsc.nasa.gov/JSCHistoryPortal/history/oral_histories/DavisHP/DavisHP_7-28-09.htm — accessed 20 March 2020).

More Information

Harold Urey and the Moon (1961)

Apollo Science and Sites: The Sonett Report (1963)

Electricity from Space: The 1970s DOE/NASA Solar Power Satellite Studies

One Space Shuttle, Two Cargo Volumes: Martin Marietta's Aft Cargo Carrier (1982)

"A Vision of the Future": Military Uses of the Moon and Asteroids (1983)

Mission to the Mantle: Michael Duke's Moonrise (1999-2009)