About a year ago, we mentioned that the European Space Agency had approved the technical concept for Reaction Engines' Skylon spaceplane. The design depends on a single engine that can work as two different kinds of rocket engines to enable runway to orbit operation, and the enabling component of this system has now been built and tested.
Rocket engines generally differ from jet engines in that rockets carry their own oxidizer, which is half of what you need to burn a fuel to generate thrust. Jet engines (which only work where there's air to breathe) suck oxygen directly out of the atmosphere, which is why planes only need to carry jet fuel. Rockets, on the other hand, carry along separate tanks of fuel and oxidizer, which is what enables the fuel to burn in space when there's no oxygen around.
Hauling an oxidizer is very expensive for rockets in terms of mass, but if you need to move heavy things quickly, there's just no way around it. But the first phase of a rocket's flight is generally through the atmosphere, where there's oxygen all over the place, and Reaction Engines has managed to invent a single engine that can use oxygen from the air until it gets up above the atmosphere, and then switches to a liquid oxidizer, saving a huge amount of space and weight and enabling single stage to orbit transport for less than half the cost of a conventional rocket system.
The reason that this is so hard to do is the speed involved: at Mach 5, air entering the engine is heated to about 2,000 degrees by compression effects, which is hot enough to melt all kinds of sensitive engine components. In order to extract safe and useful oxygen to mix with fuel, that air has to be cooled down to -200 degrees. In 1/100th of a second. Without freezing any water vapor inside the engine. The thingy in the picture above has just been tested out, and it can apparently do this.
Reaction Engines' pre-cooler uses a complex array of very small tubes with liquid helium flowing through them to suck thousands of degrees worth of heat out of air almost instantly. Oxygen from the cold air is then compressed and mixed with hydrogen to power the engine, while the helium is pumped through a liquid nitrogen boiler to cool it back down again. Not only does this proof of concept system cooling work perfectly in testing, but it's a functional piece of the final engine, and arguably, it's the hardest piece, meaning that everything is downhill from here.
The next step is to build a small-scale operational test version of the entire engine itself, and now that this technical hurdle has been surpassed, it should be much easier for Reaction Engines to get funding to do so.