LACE works by compressing and then quickly liquefying the air. Compression is achieved through the ram-air effect in an intake similar to that found on a high-speed aircraft like Concorde, where intake ramps create shock waves that compress the air. The LACE design then blows the compressed air over a heat exchanger, in which the liquid hydrogen fuel is flowing. This rapidly cools the air, and the various constituents quickly liquefy. By careful mechanical arrangement the liquid oxygen can be removed from the other parts of the air, notably water, nitrogen and carbon dioxide, at which point it can be fed into the engine as normal. The hydrogen is so much lighter than oxygen that the now-warm hydrogen is often dumped overboard instead of being re-used as fuel, at a net gain.
One issue with the LACE system is that in order to appreciably reduce the mass of the oxygen carried at launch, a LACE vehicle needs to spend more time in the lower atmosphere where it can collect enough oxygen to supply the engines. This leads to greatly increased vehicle heating and drag, which offset somewhat the savings in oxidizer weight, but this in turn is offset by higher Isp (Specific impulse) permitting a lifting trajectory which greatly reduces gravity losses. More significantly the LACE system is far heavier than a rocket engine, and the performance of launch vehicles of all types is particularly affected by dry mass, rather than any oxidizer mass which would be burnt off over the course of the flight.
The advantages, or disadvantages, of the design continue to be a matter of some debate.
LACE was studied to some extent in the United States of America during the late 1950s and early 1960s, where it was seen as a "natural" fit for a winged spacecraft project known as the Aerospaceplane. At the time the concept was known as LACES, for Liquid Air Collection Engine System, or ACES for Air Collection and Enrichment System. Both Marquardt and General Dynamics were involved in the research, and by late 1960 Marquardt had a testbed system running that was capable of running a 275 lbf (1.2 kN) thrust engine for minutes at a time. However, as NASA moved to ballistic capsules during Project Mercury, funding for research into winged vehicles slowly disappeared, and LACE along with it.
At the same time, John Scott and Bob Parkinson at British Aerospace had started some preliminary work on reusable launch systems. The two teams met and created HOTOL, which would use the BAe designed airframe with a Rolls Royce version of Bond's engine, known as the RB545. In 1986 the project was given an official go-ahead to the tune of 2 million pounds for research, but the program was later killed in 1989 when the project hit problems and the government refused further funding.
The principal designers then left to continue development on their own, but the RB545 had been classified top secret and could not be used. Instead Bond developed another version that is more advanced, known as SABRE (ostensibly for Synergic Air BReathing Engine) which is meant for their Skylon design. Funding has not been terribly forthcoming (surprising, considering the design's potential to power an SSTO craft) and development continues at a relatively low level; papers and laboratory work are ongoing. A study of an aircraft powered by a SABRE class engine is underway under LAPCAT partially funded by the EU looking towards hypersonic intercontinental travel (Brussels to Sydney in 2-4 hours non stop).