Oxygen Extraction from Lunar Samples

by Carlton C. Allen
Lockheed Martin Engineering and Sciences Co.
Carl Allen

To date, humans venturing into space have relied exclusively on equipment and supplies carried from Earth. This strategy is certainly appropriate for operations in Earth orbit, or for stays of a few days on the surface of the Moon. However, the ability to effectively utilize local resources, to "live off the land," will prove vital for long term human habitation of the Moon and planets.

Our research has focused on the extraction of oxygen, a key example of in-situ resource utilization which will directly support early human presence on the Moon. This is because one of the largest elements in any rocket is the oxygen required to burn the fuel. Locally-produced oxygen for rocket propulsion promises by far the greatest cost and mass saving of any in-situ resource.

The samples returned by six Apollo and three Luna missions contain no free oxygen, nor any water, ice, or water-bearing minerals. All lunar rock and soil do, however, contain approximately 45 wt% oxygen, combined with metals or nonmetals to form oxides. This oxygen can be extracted if thermal, electrical, or chemical energy is invested to break the chemical bonds. Over twenty different methods have been proposed for oxygen extraction on the Moon.

The reduction processes, particularly those which use hydrogen as the reducing agent, are the most technologically mature. Oxygen which is chemically bound to iron in lunar minerals and glasses can be extracted by heating the material to temperatures above 900°C and exposing it to hydrogen gas. The basic equation is:

FeO + H2 -> Fe + H2O

This process results in release of the oxygen as water vapor. The vapor must be sepa-rated from the excess hydrogen and other gases and electrolyzed. The resulting oxygen is then condensed to liquid and stored. Experiments using samples of lunar ilmenite, basalt, soil, and volcanic glass have demonstrated the required conditions and efficiency of this process.

Ilmenite - Most early work on lunar resources has focused on the mineral ilmenite (FeTiO3) as the feedstock for oxygen production. This mineral is easily reduced, and oxygen yields of 8-10 wt% (mass of oxygen per mass of ilmenite) may be achievable. Ilmenite occurs in abundances as high as 25 wt% in some lunar basalts. Maximum yields calculated as mass of oxygen per mass of rock thus range from 2-2.5 wt%.

Basalt - Previous oxygen production experiments utilized lunar basalt 70035 which was crushed but not otherwise beneficiated. The sample produced 2.93 wt% oxygen in a 1050°C hydrogen reduction experiment. Of the minerals in this rock, the most oxygen was extracted from ilmenite, with lesser amounts from olivine and pyroxene.

Soil - Oxygen can be produced from a wide range of unprocessed lunar soils, including those which contain little or no ilmenite. Figure 1 shows oxygen yield measured for four lunar soils which were reacted with hydrogen at 1050°C. The rate of oxygen extraction was highest during the first 30 minutes, but continued throughout each 3 hour experiments. Oxygen yield from lunar soils is strongly correlated with initial iron content (Figure 2). The dominant iron-bearing phases in lunar soil are ilmenite, olivine, pyroxene, and glass. Each of these phases is a source of oxygen. Ilmenite and iron-rich glass react most rapidly and completely. Olivine is less reactive. Pyroxene is the least reactive iron-bearing phase in lunar soil.

Figure 1
Figure 1. Oxygen release with time for lunar samples 74220 (volcanic class), 71131 (mare soil), 12032 (mare soil), and 62241 (highland soil), reacted in hydrogen at 1050°C.

Figure 2
Figure 2. Oxygen yield as a function of initial iron content for lunar soils reacted in hydrogen at 1050°C.

Volcanic Glass - The optimum feedstock for a lunar oxygen production process may be volcanic glass. At least 25 distinct glass compositions have been identified in the Apollo sample collection. The iron-rich species promise particularly high oxygen yields. The deposit sampled by the Apollo 17 astronauts (sample 74220) is uniformly fine-grained and friable, offering a feedstock which reacts rapidly and can be used with little or no processing prior to oxygen extraction. Complete reduction of the total FeO content of this iron-rich glass is equivalent to an oxygen yield of 5 wt%. Over 80% of this yield was achieved after 3 hours at a temperature of 1050°C (Figure 1).

The production of oxygen from lunar materials is now a reality. Oxygen release by means of hydrogen reduction has been demonstrated in the laboratory with samples of lunar basalt, soil, and volcanic glass. Yields from soils are predictable, based solely on each sample's iron abundance. The reactions are rapid, with most of the release occurring in a few tens of minutes. All of the major iron-bearing phases in lunar soil release oxygen, though with differing degrees of efficiency. These data can support the design of an oxygen production plant at a future lunar base.

For more details see: