Say at some point the human race gets its act together and actually decides to send a spaceship to another star. We’ll say it’s unmanned because that makes everything so much easier; it’s just a robot probe that’s intended to sate our curiosity about what Alpha Centauri actually bloody looks like, and so we don’t need any life support systems or generation ship malarkey going on. We’ll also say that getting the probe out of the Earth’s gravity well – aka the trickiest part of spaceflight – is similarly not an issue because we’re building it in orbit or something. This leaves us just one, fairly major problem: how do we make it go?
“But Hentzau,” I hear you say, “the laws of Newtonian mechanics say that making it go should be the easiest thing of all! After all, if something is floating in space and you give it a push it’ll just keep going forever and ever, according to Newton’s third law. Surely the hard part isn’t making it go, but in getting it to stop?” Or at least I think I hear you say that; I’m pretty sleep deprived right now so it could also be the crazed squawking of an itinerant bluebird1. Anyway, assuming a really simple case where the probe is stationary with respect to its destination you could indeed just give it a little nudge and it would start wending its merry way to Alpha Centauri without braking for anybody. The only problem is it would take a while. In fact it would take so long that I’m not entirely sure it would get there before Alpha Centauri went nova, and we don’t even have that kind of time. The human race has a notoriously short attention span and we’re going to want results soon – within the next couple of centuries, say. This means that we are going to have to mount some form of interstellar propulsion engine on our probe, and interstellar propulsion engines are, via this rather convoluted introduction, the thing I wanted to talk about today.
Making something go in interstellar space is a completely different matter to getting out of a gravity well or travelling between planets. For the former the key is a shitload of thrust delivered over a very short period of time so that you don’t fall back to Earth, which is why rocket engines are currently the best we’ve got for that. For the latter, the distances are (relatively) small enough that we have the luxury of just being able to coax something onto a transfer orbit using a very small amount of propellant and then leaving it until it eventually arrives anywhere from a few months to a decade later. These are both impractical approaches to interstellar travel for obvious reasons: rocket engines require a lot of propellant which is quickly exhausted, rendering the engine useless, while interplanetary probes might sound nippy but they are, in fact, achingly slow. For an example, the Voyager probes are the fastest man-made objects in existence, taking advantage of a once-every-175-years alignment of the planets to do some crazy gravitational slingshotting that accelerated them up to a rather bracing 15 km s-1. Let’s say we could do that again for our interstellar probe, and that we could do it in such a way that it would end up pointed at Alpha Centauri. At 15 km s-1 the probe would cover the four light years separating the Sun and Alpha Centauri in just under 80,000 years.
Clearly this isn’t practical if we want human civilisation to still exist when our probe reaches its destination, so if we intend to try interstellar travel in a serious way then we need to experiment with something a little different. Where achieving orbit is a matter of delivering a lot of acceleration quickly, and interplanetary travel relies on a small amount of acceleration delivered early, practical interstellar travel requires that we instead subject our probe to a small amount of acceleration over a very long timespan. The thing with velocity in space is that it’s cumulative. There’s no air resistance dragging you back; once you add 5 km s-1 to a spacecraft’s relative velocity that 5 km s-1 will stay there forever until you remove it by making a burn in the opposite direction and manually decelerating the spacecraft. An engine that provides an acceleration of 2 cm s-2 might not sound very good, but if it can provide that rate of acceleration for a full year the spacecraft it’s attached to will end up travelling at a whopping 630 km s-1, covering the distance to Alpha Centauri in a nippy 452 years. And that’s if it stops accelerating after a year, remember. If it keeps going there’s no reason it couldn’t reach a significant fraction of light speed.
There’s only one problem with this approach. Well, two problems, with the second, smaller one being that you can only accelerate like this for the first half of the trip. Then you have to turn the spacecraft around and burn the engine in the opposite direction to slow it down and make sure it doesn’t whip by its target at this unholy velocity you’ve accelerated it to. However, the big problem is that providing sustained low-grade acceleration for a prolonged period of time is impossible using conventional rocket engines. They’re built for rapid acceleration over small time scales and spend their propellant profligately in achieving this end, and even if you burn it as efficiently as possible a rocket engine would still run out of fuel long before it ever achieved any significant velocity. You can’t stuff a half-century’s worth of rocket fuel onto a spaceship, after all.
In order to do interstellar travel, then, we’re going to need a whole new type of spaceship engine. The thrust it provides does not need to be large, but it does need to be constant and very long-lasting. This gives us two options:
1) Use a very efficient fuel source that will somehow last the duration of the trip. This basically means nuclear rockets, a concept that’s been on the drawing board for years but which has never gone anywhere because people are a bit nervous about having nuclear reactors hurtling through the sky at several hundred miles per hour. There are high-efficiency thrust systems currently in development – see ion thrusters and the SMART-1 probe – that nevertheless require a powerful source of electricity to run. Nuclear reactors fit the bill and would last a half-century or more to boot, but they require several quantum leaps on the technology side of things (like the LFTR reactors I was talking about the other day) before they’re viable as a propulsion system.
2) Don’t carry your fuel with you. Get it from somewhere else, while you’re travelling. This is not as dumb as it sounds, with the two major ideas in this category being light sails and the Bussard ramjet.
Light sails are fairly easy to understand. All light exerts an infinitesimal force whenever it strikes an object, kind of like a really crappy version of the wind on Earth, so you just put a really big solar sail on your probe and focus an incredibly powerful laser on it from the Earth which “pushes” the spacecraft along. This sounds like fun, but in reality the power of the laser would attenuate massively as the spacecraft got further away from Earth necessitating a planet-sized sail in order to focus the light properly. This is slightly impractical for obvious reasons, and while there are designs which use microwaves which should be more efficient it’s going to be some time before we’re capable of building the size of spacecraft an interstellar-capable solar sail would demand.
Then there’s the Bussard ramjet, a concept which is quite popular in “hard” sci-fi novels. Remember the atmospheric ramjets I talked about in the Skylon post? This is like that, except instead of using the Earth’s atmosphere as reaction mass the Bussard ramjet instead has two huge electromagnetic field generators which it uses to scoop up material from the interstellar medium and compress it until it achieves fusion, generating energy which is used to propel the spacecraft. This relies on the fact that interstellar space isn’t quite empty. It actually has quite a lot of stuff in it, it’s just that that stuff is spread out so much that there’s maybe an atom’s worth of it per cubic centimetre. While it’s certainly theoretically possible, though, the Bussard ramjet has even more question marks hanging over it than the light sail, including some potential problems which are similar to the ones atmospheric ramjets have to deal with (i.e. drag forces). Finally, at the end of the day the Bussard ramjet is just a fancy version of a fusion rocket, and if we can do that we might as well just concentrate on option one.
For our short range trip out to Alpha Centauri I think nuclear rockets are probably the most feasible propulsion system available to us. The journey duration could well be short enough that a nuclear reactor would remain viable for the entire length of the trip, as well as providing power for the spacecraft once it finally got there. For longer distances, though, we’d have to start looking at the radical departures from regular spacecraft design seen in approach 2). It’s simply not possible to propel and power a spacecraft that far while carrying your fuel source with you, so it needs to come from some external source. So while they may sound kind of ludicrous, if interstellar spaceflight has a future it probably lies in concepts like the Bussard and light sails. Of course they look kind of silly, but then spaceflight isn’t particularly big on dignity anyway.
1. Fun fact: bluebirds don’t actually nest anywhere in Dover because they’re an American species, thus proving that the people who write lyrics to songs are imbeciles who are more fond of metaphor than they are scientific accuracy. I hate songwriters.