The entry says that the men in spacecraft 012 sat in pure oxygen at 16.7 pounds per square inch because of a decision nobody thought of as a decision. This piece sets out where that atmosphere came from, as the Review Board’s design panel reconstructed it in 1967, what NASA had published about fire in oxygen before the accident, and why, after the fire, the Board and the Senate let the flight atmosphere stand and changed only the pad.

Mercury, 1960

A spacecraft’s atmosphere has to do several things at once. It has to keep the crew conscious, which is a matter of the partial pressure of oxygen; it has to spare them the bends if the cabin loses pressure, which argues against nitrogen; it has to be light, simple and reliable, which argues for a single gas; and it has to let a man in a pressure suit move, which argues for a low total pressure, since the suits of the 1960s stiffened above about 3.8 pounds per square inch of difference between inside and out. Project Mercury settled on pure oxygen at five pounds per square inch in flight, and a NASA Life Sciences Committee under W. Randolph Lovelace II approved it.

The pressure on the pad was settled by an accident. On 21 April 1960 a McDonnell test pilot, G. B. North, climbed into a boilerplate Mercury capsule in an altitude chamber at the company’s plant for the first manned test of the environmental control system. He had breathed oxygen beforehand and his suit circuit had been purged; the cabin had not, because the design relied on the climb to orbit to flush four-fifths of the air out through the relief valve. The chamber was pumped down to the equivalent of 27,000 feet. After about forty minutes at five pounds per square inch the test was stopped because North was unconscious. The Board’s panel gave the cause in a sentence: “leakage of nitrogen from the spacecraft air into the pressure suit circuit had gradually decreased the partial pressure of oxygen below physiologically acceptable limits.” Parts of the suit circuit ran at a lower pressure than the cabin, and the cabin’s nitrogen had crept in. Three more manned tests, two on 2 June and one on 6 June 1960, were stopped for the same reason.

The fix was to take the nitrogen away before launch. “As a result of these incidents,” the panel wrote, “the prelaunch procedure for all Mercury spacecraft, both astronaut and chimpanzee, was changed to require that the cabin be purged with oxygen prior to launch.” And then: “The requirement for purging the cabin with pure oxygen at approximately 15 psia during the prelaunch period of several hours has been continued for all manned spacecraft launched in this country. This same procedure has been used also on all manned spacecraft vacuum chamber tests in the Mercury, Gemini and Apollo Programs.” Gemini took the Mercury atmosphere unchanged, and its environmental control system “performed successfully in 10 manned flights, and in a large number of manned and unmanned altitude chamber tests and prelaunch operations.”

Apollo, 1961 to 1962

Apollo’s flights would be longer, and in 1961 the physiologists recommended something different: a mixture of oxygen and nitrogen at seven pounds per square inch. The reason, the panel recorded, “was concern by physiologists that two-week Apollo missions in a 5 psia 100% oxygen environment (used in the Mercury Program) could cause pulmonary atelectasis (collapse of lung tissue)”, which had been seen after long periods of breathing pure oxygen. Against it stood the bends, if a mixed-gas cabin lost its pressure quickly. A test programme was run on both questions. It found that three hours of breathing oxygen before launch would prevent the bends if the cabin decompressed during or just after it, and that two weeks in pure oxygen at five pounds would not collapse anyone’s lungs. “Based on the results of this test program,” the panel wrote, “NASA decided in 1962 that the Apollo spacecraft would also use the 5 psia 100% oxygen atmosphere used in the Mercury and Gemini Programs.” It listed six things the choice bought: continuity with Mercury and Gemini, no bends, simple and reliable hardware, a cabin in which the crew could work in shirtsleeves, few restraints on going outside, and the most mobility the suits allowed.

Nothing in that reasoning was about the ground. The five pounds of oxygen in flight was the subject of the analysis; the fifteen or sixteen pounds on the pad was a procedure inherited from 1960, and its purpose was to keep nitrogen out of the suits. The Board’s panel on design, re-examining the decision in 1967, held that “Previous analyses leading to the decision to use 5 psia pure oxygen cabin environment in space are still valid”, that a two-gas system would need duplicate controls and a way of measuring the partial pressure of oxygen, and that “These additions introduce additional crew-safety failure modes into the flight systems.”

What had been published

In 1964 NASA’s Office of Manned Space Flight published, as Special Publication 48, a review of the literature on fire and explosion in space cabins, written under contract by Emanuel M. Roth of the Lovelace Foundation in Albuquerque. It was the second part of a four-part study of cabin atmospheres; the first was on oxygen toxicity, and the fourth, still to come, on one gas against two. The acting director of space medicine, George M. Knauf, wrote in the foreword that the book “provides a readily available summary of the open literature in the field” and was “intended primarily for biomedical scientists and design engineers”, and that “for scientist and engineer alike it is anticipated that this study will become a basic building block upon which research and development within the space community may proceed.”

Roth’s chapters ran through ignition, flame propagation, detonation, extinguishment and the flammability of fabrics, fluids and solids. On sparks: “The generation of an electrostatic spark within a space-cabin system is always a danger.” On the effect of oxygen: in going from air to pure oxygen “the minimum required spark energy decreases by 3 orders of magnitude”, and the range of mixtures that will burn widens from a narrow band to most of the scale. On what happens once a fire has started: “Once burning is started in a pure oxygen atmosphere, however, it proceeds much more vigorously and at higher temperatures. Flames are blue-white in contrast to the orange and yellow flame in normal air.” Two of his figures on spark ignition were drawn after the work of Robert W. Van Dolah and his colleagues, the explosives specialist who three years later sat on the Review Board. The Senate’s committee cited the book in a footnote to its finding that “NASA itself had commissioned research on the cause and effects of fires in oxygen-rich atmospheres.” It cost a dollar from the Government Printing Office.

What the organisation weighed against it was experience. Webb put the experience into his statement of 25 February 1967: Mercury and Gemini, “and a number of instances where a clearly positive source of ignition did not result in a fire. In one such instance an electric light bulb was shattered, exposing the incandescent element to the oxygen atmosphere without starting a fire.” The Board’s own explanation of why the materials in the cabin had been thought safe was that each had been tested on its own, in small samples; North American’s was the same: “What was not fully understood by either North American or NASA was the importance of considering the fire potential of combustibles in a system of all materials taken together in the position which they would occupy in the spacecraft and in the environment of the spacecraft.”

What the Board kept

After the fire the panel measured what a diluent would have bought. Tests with materials in oxygen at sixteen pounds and at five found the downward burning rate only 1.3 times faster at the higher pressure, and against the most nitrogen the body would tolerate, 3.5 pounds of each gas, the ratio was 1.8. “It must be concluded,” the panel wrote, “that burning rates of materials are significantly reduced only when large amounts of diluent are used. The limited quantity of diluent acceptable by physiological criteria contributes very little to the reduction of burning rate over that in pure oxygen.” Its finding was that flammability “is varied by only a factor of 3 or 4” by the diluents an atmosphere could hold, and its determination was to keep the flight atmosphere. The Board’s main report said the same more cautiously: a diluent “would undoubtedly reduce the risk of fire to a significant degree” and would “introduce other operational problems and risks”, and the technology for controlling two gases “is still far from being fully developed.” It recommended that the studies go on.

The Senate committee recorded where NASA had got to by January 1968: the flight atmosphere “should continue to be 100 percent oxygen at 5 p.s.i.a.”, but the command module had been modified “to be capable of using air, as well as pure oxygen, as a pressurant on the launch pad”, and the committee added the caution that “the dual gas cabin atmosphere, while reducing the fire hazard, creates other risks such as the risk of the astronauts getting the ‘bends’ if their cabin pressure is reduced quickly.” The Kennedy historians record the outcome: after a year and a half of testing NASA settled on sixty per cent oxygen and forty per cent nitrogen for the pad, with the nitrogen replaced by oxygen after launch. Atwood of North American said that the biggest mistake had been to lock the crew in and pump in oxygen above sea-level pressure. The materials were changed everywhere; the atmosphere was changed only where the fire had been, on the ground, which is where the sixteen pounds had never had a reason to be except to keep the air out.

Sources

• Apollo 204 Review Board, Appendix D, Panel 9 (Design Reviews), enclosure 9-2, “Spacecraft Atmospheres”: the Mercury chamber tests of 1960, the purge requirement, Gemini, the Apollo recommendation of 1961 and decision of 1962, and the panel’s measurements and findings on diluents. nasa.gov
• Apollo 204 Review Board, Report of Apollo 204 Review Board to the Administrator, 5 April 1967, Part V, “Spacecraft Atmosphere”, and Finding 9; and James E. Webb, statement of 25 February 1967. nasa.gov, Webb
• Emanuel M. Roth, Space-Cabin Atmospheres, Part II: Fire and Blast Hazards, NASA SP-48, 1964. ntrs.nasa.gov
• US Senate, Committee on Aeronautical and Space Sciences, Apollo 204 Accident, Report No. 956, 30 January 1968, “One Hundred Percent Pure Oxygen Environment”. nasa.gov
• Courtney G. Brooks, James M. Grimwood and Loyd S. Swenson, Chariots for Apollo, NASA SP-4205, 1979, chapter 9, for North American’s statement and Atwood’s view; Charles D. Benson and William Barnaby Faherty, Moonport, NASA SP-4204, 1978, chapter 18, for the sixty-forty mixture. Chariots, Moonport