Lightspeed: Edited by John Joseph Adams




Engines for the High Frontier

“The engines are givin’ ye all they can, Captain!” (Scotty, approximately, in many Star Trek Episodes.)


We want engines to get us into space and take us to the stars. Of course, as many note, we aren’t quite where we want to be yet. But there is hope.

At the moment, spaceship engines can be classed into three categories: rockets, sails, and “other,” and each works in their own, individual way. Rockets work by pushing something out the rear; reaction equaling action, you go in the other direction. With sails, something external pushes. And in the “other” category are things like “space drives” and ramjets. But more about those later. For now, let’s start with rockets.

Self-contained rockets use on-board energy to push out on-board propellant. The energy may be in the fuel itself, or may come from some other kind of on-board energy source, such as a nuclear reactor. How fast such a rocket goes depends on how fast it can expel mass (its exhaust velocity), and for how long (its energy supply).

To go as fast as its own exhaust velocity, a rocket needs to carry about 2.7 times its empty mass in fuel, its “mass ratio.” To go twice as fast, square that (7.4), three times as fast, cube it and so on. Five times as fast requires a mass ratio of 148.

Now, to package 148 times as much fuel in a rocket as the rest of it weighs is going to take some clever thinking. For instance, the fuel can be some kind of solid that doesn’t need a tank, while the payload can weigh almost nothing (something that will be possible in the coming era of nanotechnology). The engines, of course, will have to weigh something because of the power they need to handle, but you can always discard them or grind them up and use them as fuel as you go along.

Well, maybe.

For practicality’s sake, let’s take three times the exhaust velocity with a mass ratio of twenty as our speed limit for self-contained rockets, admitting a bit of wiggle room.

Practical chemical fuels have exhaust velocities of around 3 to 5 kilometers per second, and it takes about 10 km/s to get into orbit. People have been trying for some time to build single stage to orbit rockets, and not quite succeeding; both the funding and engineering problems are huge. Until those are overcome, we’ll have to rely on practical designs which have two stages. One of the most practical is the SpaceX Merlin engine, which achieves reliability and performance by having single hardware systems perform multiple functions to reduce weight and complexity.

But that’s a very near term thing. What we want is to go to the stars! We want engines that will push us so close to the speed of light that time slows down relative to that of the stars and the universe outside looks compressed, with the light ahead visibly shifted toward the blue and the light behind toward the red. At 87% of the speed of light, your apparent “map” velocity (star map distance covered) divided by the time elapsed in the spaceship would be twice the speed of light! Of course, the clocks where you arrive would tell a different story, but that’s relativity, and a whole other thing.

Now, to achieve such speeds requires new kinds of engines, not to mention fuels.

In space, a big mass ratio would be easier, but still fifteen to twenty kilometers per second is pretty much all you’ll get from chemical rockets—Alpha Centauri in a few thousand years. Nuclear fuels are better. There are lots of issues and little space here, so let’s just say that we’re looking at maximum exhaust velocities of around 0.08 c. For a spacecraft to achieve 0.9 c, even those “mass ratio of 148” tricks won’t do it.

Antimatter, if you could make it in quantity, (and that’s a very big if) ends up being much like a better nuclear fuel. But you don’t get “pure energy” from the annihilation process, you get a lot of hard radiation, only some of which can be captured for propulsion purposes and exhaust velocities might be up around 0.4 c. You can wave your hands at 0.9 c with antimatter.

But how about we banish energy limits altogether? There’s a class of rockets that gets its energy from elsewhere—the Sun, laser beams, or microwaves, and in these cases, there are no limit (in principle) to the amount of energy available per unit fuel mass, though, of course, there are still practical limits to the power, thrust and acceleration.

In the context of propulsion, solar energy isn’t very powerful, so solar rockets won’t accelerate rapidly. Beams can be much more powerful, but there are also engineering details involved in focusing the beam on the rocket over long distances.

However, in the end, the ultimate energy source might just be the pulsating fabric of spacetime itself; alas, there are as yet no good ideas for tapping into that.


Now, there are many, many devices for pushing stuff out the back end of a rocket. For example, expansion nozzles (material or magnetic) which turn a hot blob of gas or plasma into a directed flow of matter rearward. There’re also ion rockets which use electric fields to accelerate charged matter. However, since like-charged particles (atoms, molecules, clusters, dust…) repel each other it’s hard to get an exhaust dense enough to provide a lot of thrust, even if you have the energy.

Plasma (a mixture of positive and negatively charged particles, generally too hot to recombine into a neutral gas) can be accelerated by magnetic fields, and because plasma is neutral overall, a plasma exhaust can be compressed enough to provide more thrust than ion exhausts. (For a future plasma engine that’s received lots of attention over the years, Google “VASIMR.”)

Then there are particles or pellets that can be accelerated rearward by various kinds of electromagnetic guns to provide thrust, and the many mechanisms for doing so.

Photons (light rays) can be shot out the rear as well; however, they don’t provide much thrust per photon, so using enormous power supplies to produce the push of a mouse fart seems rather silly. And just for the record, while a photon exhaust may seem to be mass free, it really isn’t. From Einstein’s E = mc2, to generate each photon requires a mass of m = E/c2. In other words, photon rockets have mass ratio issues just like other rockets…as well as having really wimpy thrust.

And finally, there’s the “space drive” which purports to somehow use the fabric of space itself as the reaction mass. That would be a nice trick, but the physics of it is still very iffy. And just like the photon rocket, it would run on energy, which has mass, so, once again, no free launch there.


Beyond rockets, there are sails.

Sails operate by reflecting photons, particles, or both for free thrust (well almost free—the sails have some mass). Now, if they rely on natural “wind” sources, like the Sun, they accelerate very slowly. However, if we provide the wind, say by using robots to build thousands of beam projectors and sun-power stations to energize them, then, in principle, there’s almost no limit, and sails, in principle, can approach the speed of the beam that pushes them. But while light beams are easier to make, in the long term, particle or pellet beams may be preferred as, just like the photon rocket, they deliver much more momentum for the energy used to propel them.


And finally we come to the “other” category, also known as the “don’t hold your breath” category.

First off, let’s talk briefly about the Bussard ramjet, which is being mentioned here simply for historical reasons. The Bussard ramjet was designed to use interstellar hydrogen as both a fuel and a reaction mass. Unfortunately, there’s really not enough interstellar hydrogen per unit volume to make this work well, and the drag on the scoop needed to collect the hydrogen severely limits its ultimate velocity. However, despite these drawbacks, people continue to work on variations of this idea, so who knows?

And as for warp drives and their like, those engines are mainly the province of fiction. And while there is some physical basis in general relativity for a few of these, the energies required to power them are truly astronomical—as in the energy equivalent of the masses of small planets astronomical. And just to up the impossibility factor, any system that would deliver a payload or message faster than light would also be a time machine, and so lead to causality paradoxes that many think nature will not allow. So as we said, don’t hold your breath.

For even the best of the above concepts, there are, as always, serious engineering issues. Future engineers, however, should eventually be able to solve enough of them to allow us, at long last, to reach the stars.

Ad astra!

Enjoyed this article? Consider supporting us via one of the following methods:

Gerald D. Nordley

Nordley, GeraldGerald D. Nordley is an astronautical engineer and writer of over 50 published works of nonfiction and short fiction, two novels, and a Mars-related story collection, After the Vikings from His main interest is human space exploration and settlement and he tries to keep his scientific background real.  He has four Analog “Anlab” readers’ awards, and a Hugo and a Nebula nomination.  His latest novel, To Climb a Flat Mountain, appeared in Analog as a serial in the Nov. and Dec. 2009 issues. He lives in Sunnyvale CA with his wife, a retired Mac computer programmer