Rockets suck. This is a thing that we have established here; they’re terrifically awful ways of getting into space that are only used because nobody has really come up with anything better. There’s all sorts of ideas for wacky drive systems once your spacecraft is actually in space – ion drives, solar sails, Bussard ramjets – but these all sidestep the real problem facing future space travel, which is that you have to get out of the Earth’s gravity well first. This is not easy; even though the Earth is pretty small for a planet it’s still the heaviest of the four terrestrials and has what is to us a very hefty gravitational pull.
In order to exit Earth’s gravity well, an object being launched on a ballistic trajectory from the Earth’s surface must attain escape velocity. Escape velocity is calculated by
where G is the gravitational constant, M is the mass of the Earth and R is the radius of the Earth. For “ballistic” you can read “in freefall” – i.e. an object moving without any other forces acting on it. Rockets do have external forces acting on them thanks to their propulsion systems, so a rocket doesn’t necessarily have to reach escape velocity of 11.2 kilometres per second in order to escape the Earth’s gravity so long as it’s under constant power. However, the amount of energy it will have to expend will be the same as this hypothetical ballistic object, and that can be worked out by using the equation for kinetic energy.
where m is the mass of the spacecraft. With Earth’s escape velocity being 11.2 km s-1, we can rearrange this equation to find out how much energy per kilo of mass a launch vehicle will need to have to escape Earth’s gravitational pull, and this turns out to be about 63 megajoules per kilogram. I always get a little lost when trying to visualise what a joule is in everyday terms, but Wikipedia tells me that 63 megajoules is roughly the amount of energy you’d release if you detonated fifteen kilograms of TNT. So for a crude method of visualising this, go and look at a picture of a rocket. Imagine a pile of TNT fifteen times as big. Blowing up this pile of TNT would release the same amount of energy as would be needed to get that rocket out of the Earth’s gravity well.
(One significant caveat here: rockets going into orbit don’t need to expend this much energy. They just need to go fast enough that they never hit the ground.)
This amount of energy is fixed. There is no way of getting around it; if you want to go into space, you have to expend at least 63 MJ per kilo of spacecraft mass and this energy has to come from somewhere. For rockets, it comes from rocket fuel; unfortunately rocket fuel increases the mass of the rocket, which requires more rocket fuel, which increases the mass of the- actually I think I’ve done this already, haven’t I. Anyway, rockets suck, but we still use them because the alternative solutions are, to put it mildly, just a little bit far-fetched.
Solution 1: Build better rockets.
Ahaha. Ahahaha. Ahaha. No. Rocket science isn’t necessarily all that complicated – I can understand it, after all – and one of the things that struck me when I was studying space science in the second year of my undergraduate degree was that rockets are a dead end. Well, that’s a bit harsh. Better to say that they’ve been refined as much as they can be.
Two things affect the amount of thrust you can get out of a rocket: the expansion velocity of the exhaust gas you’re creating by burning the fuel, and the size of the rocket nozzle. Rapidly expanding rocket fuels tend to be rather unstable – again, see the Nedelin catastrophe – and so there’s an upper limit on just how volatile you want to make that. What about the rocket nozzle? How wide/narrow this is determines the pressure at which the exhaust gases exit the rocket. For reasons I won’t go into, rockets produce the maximum amount of thrust when the pressure of the exhaust gas equals the pressure of the ambient atmosphere through which the rocket is flying. Atmospheric pressure decreases as you ascend into space, so ideally you want a rocket nozzle that automatically adjusts in size with altitude to keep the rocket working as efficiently as possible. This has been done. After that, there’s no real way to further improve a chemical rocket. We’ve taken them just about as far as we’re going to.
Solution 2: Don’t launch them from Earth.
One of the attractive things about a Moon colony – aside from the whaling opportunities – is that the Moon has a gravitational pull less than one-sixth that of Earth’s. It has an escape velocity of 2.38 km s-1, and consequently it would only take 2.83 MJ to get one kilogram of mass out of the Moon’s gravity – under a twentieth the energy you’d need to get the same kilo off of Earth! Probably the best way to illustrate this is the ascent module on the Apollo lunar landers; that little capsule was able to create enough thrust to return to orbit to rendezvous with the command module.
It would be far easier, therefore, if we could somehow launch rockets from the Moon instead. Unfortunately this idea merely replaces the problem of getting out of the Earth’s gravity well with the even larger problem of building the necessary industrial base on the Moon to build and launch rockets. I’m not saying it couldn’t be done, and if somebody did manage to do it there’s every chance we’d see the commonplace spaceflight depicted in movies like 2001 (if you were lucky enough not to live on Earth, anyway), but the level of commitment and resources it would require would be staggering.
Solution 3: Mass drivers.
This one is simple, and operates on the principle that while achieving 63 MJ per kilo is a tall order for a chemical rocket, if we could somehow use electricity instead it’d be far easier to generate the energy required. The idea is you have a long, long, long, long launch track, kind of like a very high-tech version of Japan’s bullet train network (see maglevs for further information), and you accelerate the thing you want to launch down this track using superconducting electromagnetism. The track slopes upwards towards the end, and so once it reaches the end of the line your launch vehicle is shot into the stratosphere. At this point some small rocket boosters would take over and move the launch vehicle to the desired orbital trajectory. Building this thing would be a bit more feasible than the Moon colony, but there’s just one tiny snag: room temperature superconductors capable of carrying the currents required haven’t been invented yet.
Solution 4: Project Orion.
In which the Orion spacecraft is supposed to fart out shaped nuclear explosions, the brunt of which is directed against an impact plate on the ass of the spacecraft which makes the nukes “push” the spacecraft along. Utterly, utterly mad.
Solution 5: A space elevator. Also unicorns.
A very old concept that’s rather in vogue in 4X games, for some reason. As mentioned here geostationary satellites orbit the earth with a period of one day – that is, they remain above the same point relative to the Earth’s surface throughout their orbit. In theory, if you had access to a material strong enough as well as the combined and focused resources of the largest superpowers on Earth over a half-century or more, you could lower a cable from geostationary orbit all the way down to the surface of the Earth and simply have vehicles ride up and down the cable just like an elevator car. This is attractive because the energy costs are similar to those of a mass driver launch with the added bonus that you get most of it back as elevator cars come back down from orbit. If we then wanted to launch stuff further out into the solar system we could take advantage of the fact that the space elevator would kind of act like a giant sling to any payload launched from the far end.
Sounds fun, right? Perhaps so, but space elevators have several minor niggles past the huge advances in space technology, robotic manufacturing, materials science and megastructure construction (I just made that last one up, but someone’s going to have to invent it before they can build the elevator) that would be required, not to mention the totalitarian world government that’d have to be in place to keep everyone pointed in the right direction long enough to finish the bloody thing.
- In order to make it work you need a big counterweight of some kind to produce the necessary centrifugal force to keep the elevator cable taut. These days space elevator designs merely pay out the cable a little further past geostationary orbit to provide a counterweight mass rather than the previous idea of moving an asteroid to GEO to act as counterweight, but even if you do this you still need the asteroid because a) you need to build the elevator from both ends at once and so you’re going to need a base of operations in space from which to do it, and b) an asteroid would provide necessary raw materials for cable construction. So first you need to move several million tonnes of rock into geostationary orbit. This is just a little bit tricky to do, and I imagine it might make people down on the surface a little bit nervous as well what with the potential consequences if something goes wrong.
- There is no material which can currently be mass-produced in the quantities required which has the necessary tensile strength to support thousands of kilometres worth of its own weight. People always point to carbon nanotubes as the catch-all solution, but while they have the theoretical capacity to support the quantity of mass required, it would require the cable to be structurally flawless over its entire 40,000 km-odd length. Even one flaw would introduce a weakness which could be potentially fatal to the whole shebang.
- Transmitting power to the elevator cars is also going to be a bit of a bugger, since you still need that 63 MJ/kg to get into orbit. A nuclear power source would do it, but that’d probably make passengers a bit uncomfortable. Solar panels would increase the weight of the car, while using the cable itself to transmit power is going to run into the very inefficiency problems the space elevator is supposed to avoid. The current favoured proposal is wireless energy transfer via lasers or something, which has the tiny drawback that it is literally space magic.
- Elevator cars would travel fairly slowly, with an ascent to GEO taking anywhere from 6-12 hours to a full day. This brings the Van Allen radiation belts into play; the cars would travel through them slowly enough that anyone inside would end up taking lethal doses of radiation. It would be necessary to shield the cars in some way, which again would add to the weight.
Personally I think the space elevator is the ultimate manifestation of Archimedes and his lever; something that is theoretically possible but practically stupid. We can’t even agree on the best way to fly three people up to LEO; we abandoned the moon after sending barely two dozen guys up there. We are nowhere near being ready to take on the incredible engineering challenges of building something like this, and so anyone who thinks the space elevator is going to be built in the next half-millenia needs their head examined.
Where do I think the likely future of spaceflight lies? Well, assuming the world doesn’t enter a period of terminal dystopian senescence in the next couple of decades I think the mass driver concept is probably the most feasible solution as long as somebody comes along to solve the superconductor problem. It’s the smallest engineering project, it has the very great advantage of the thing being built on Earth rather than in space or on another planet, and it has the fewest question marks over necessary technology advances. Plus, if aliens ever try to invade we can use it to shoot rocks at them. Welcome to Earth.