The US Air Force's aerospaceplane project encompassed a variety of projects from 1958 until 1963 to study a fully reusable spaceplane. A variety of designs were studied during the lifetime of the project, including most of the early efforts on liquid air cycle engines (LACE) and even a nuclear-powered ramjet.
The effort was started largely due to the work of Weldon Worth at the Wright-Patterson AFB, who published a short work outlining a manned spaceplane. AF officials were interested enough to start SR-89774 ("SR" stands for "study requirement") for a reusable spaceplane in 1957. By 1959 this work had resulted in the Recoverable Orbital Launch System, or ROLS, based around a LACE engine, known at the time as a Liquid Air Collection System, or LACES.
Further work showed that more performance could be gained by extracting only the oxygen from the liquid air, a system they referred to as Air Collection and Enrichment System, or ACES. A contract to develop an ACES testbed was placed with Marquardt and General Dynamics, with Garrett AiResearch building the heat exchanger for cooling the air. The original ACES design was fairly complex; the air was first liquified in the heat exchanger cooled by liquid hydrogen fuel, then pumped into a low-pressure tank for short-term storage. From there it was then pumped into a high-pressure tank, where the oxygen was separated and the rest (mostly nitrogen) was dumped overboard. In late 1960 and early 1961 a 125 N demonstrator engine was being operated for up to 5 minutes at a time.
In early 1960 Air Force offered a development contract to build a spaceplane with a crew of three that could take off from any runway and fly directly into orbit and return. They wanted the design to be in operation in 1970 for a total development cost of only $5 billion. Boeing, Douglas, Convair, Lockheed, Goodyear, North American, and Republic all responded. Most of these designs ignored the ACES system and instead used a scramjet for power. The scramjet had first been outlined at about the same time as the original LACES design in a NASA paper of 1958, and many companies were highly interested in seeing it develop, perhaps none more than Marquardt, whose ramjet business was dwindling with the introduction of newer jet engines and who had already started work on the scramjet. Both Alexander Kartveli and Antonio Ferri were proponents of the scramjet approach. Ferri successfully demonstrated a scramjet producing net thrust in November 1964, eventually producing 517 pounds-force (2.30 kN), about 80% of his goal.
Later that year a review suggested that the basic concepts of the aerospaceplane were far too new for development of an operational system to begin. They pointed out that far too much was being spent on development of the aircraft, and not nearly enough on basic research. Moreover, the designs were all extremely sensitive to weight, and any increase (and there always is some) could result in all of the designs not working. In 1963 the Air Force changed their priorities in SR-651 and focused entirely on development of a variety of high-speed engines. Included were LACES and ACES engines, as well as scramjets, turboramjets and a "normal" (subsonic-combustion) ramjet with an intake suitable for use up to Mach 8. In October a further review concluded that the technology was simply too new for anyone to predict when any such aerospaceplane could ever be built, and funding was wound down in 1964.