This question was inspired by some comments of mine I left on an answer here:

Is it better to develop more powerful rockets instead of seeking and developing new technologies?

and thus in turn both the OP's question about how to aim a research career (presumably) toward the development of improved technologies for space propulsion and the answer and comments to that question that suggested that we didn't necessarily need to go so far as inter-stellar travel, and with new physics, and that improved inter-planetary would still be very good, and possible with "existing physics only", which led me to point out that there may still be problems even with interplanetary travel if you want it to be fast enough that ended up in questions that probably merit a thread all their own and so here it is.

In particular, I was curious as to what kind of technology might be required if, instead of interstellar travel, we are going to limit to fast interplanetary travel but with a max transit time of 30 megaseconds(**) (a bit under one Earth year, or half a Martian year) to at least get as far as Pluto(*), at about 6500 Gm distant from the Sun, though in the comments made accommodative room to envision in more detail a perhaps slightly milder scenario of transit to Mars with an average trip time of 2 Ms (~23 d), with 225 Gm average distance (for comparison, current missions average around 20 Ms (0.65 year), so this is aiming to make transit a full 10 times faster).

And I gave some calculations involving dividing that into a 500 ks constant acceleration phase, 1 Ms cruise phase, and 500 ks constant deceleration phase to end with Mars arrival. From simple geometry of the areas under the curves, one can see that if one makes the total acceleration/deceleration time equal to the cruise time, $t_c$, then the distance one travels as a function of cruise time and cruise speed $v_c$ is

$$D(t_c, v_c) := 1.5v_ct_c$$

which can be inverted to find $v_c$ as a function of $t_c$ and a goal distance $D_\mathrm{goal}$, which if we take as 225 Gm and a $t_c$ of 1 Ms gives a $v_c$ of 150 km/s as being necessary. Note that this is rather rough since it excludes gravitational effects in the Solar System - I'm approximating as a straight shot since we're well above the usual astrodynamic speeds. Since you have to speed up and slow down, that means a total mission budget of 300 km/s of delta-V for one such transit, and an acceleration of $0.30\ \mathrm{m/s^2}$ or about $0.03\ \mathrm{gee}$.

And moreover, if you imagine a 100 Mg (megagrams, same as tonnes) craft as being propulsed, that means about 30 kN of continuous thrust, or 3 tonnes of gravitational equivalent, will be required for a 500 kilosecond period, twice, not accounting for any rocket fuel usage involved.

If we wanted to set the goal being to develop a spacecraft capable of that, what would be the most viable option a putative researcher might try for to contribute thereto? Heck, even throwing out the 300 km/s target, what about just 100 km/s? Clearly, by the tyranny of the rocket equation, chemical rockets won't work, but 30 kN, while much less thrust then a CR, is also a lot more thrust than many proposed alternatives, e.g. ion drives and light sails.

As an example of that last point, to get 30 kN out of a light sail, you need an incident power by $F = \frac{P}{c}$ thus $P = cF$ of about 9 TW, or 9000 GW. At a solar constant of roughly $1\ \mathrm{kW/m^2}$ which becomes $1\ \mathrm{GW/km^2}$, you need a $9000\ \mathrm{km^2}$ light sail, a fair bit larger than the size of the US state of Delaware and thus clearly a colossal engineering project (what's the word on that being viable to launch sometime during the next century / 3 Gs, or even just the remaining lifetime of an aspiring or beginning college student of typical age of around 18 years (570 Ms) at the present time, assuming no radical longevity breakthroughs?) and even if you use beamed power to cut the sail size, a 9 TW array could be used for interstellar unmanned probe missions and thus we're well in the overlap range with interstellar travel already, 100% defeating the idea of concentrating on one over the other.

Of course, we could relax time, but we still want it to be significantly faster - as in whole-number multipliers of speed greater than 1 at least - than existing chemical missions.

So with that in mind, my final question is, and just scaling even further to a smaller speed point: is there any feasible research path toward getting, say, mission delta-vee of at least 100 km/s, or 50 km/s cruise speed on a single fueling (or even external power), with perhaps a 1 Ms rise and fall time(XXX), for a craft capable of transporting humans?


(*) YES, I'm aware that the "official" definition of a planet excludes it, to me it is a planet, I'm with many on the New Horizons team on that, and if that means we need to admit more to the "club" like Ceres, Eris, etc. to maintain cogency, then the more planets, the merrier, pillory me all you want :g:

(**) 1 Ms, 1 000 000 s, equates to 11 days and 49.6 kiloseconds.

(XXX) That's meant to be 1 Ms each, for ramp-up/ramp-down individually, for this technology checkpoint, so again, more relaxed, instead of the 0.5 Ms in the more optimal scenario.

  • 1
    $\begingroup$ Your advocacy of the SI metric system stands in the way of communication. I've stopped reading on the second paragraph. $\endgroup$ May 5, 2019 at 13:44
  • $\begingroup$ @Diego Sánchez : Added some figures in 'conventional' days/years for the principal time periods involved. $\endgroup$ May 5, 2019 at 13:59
  • $\begingroup$ Expecting to see nuclear (nuclear-pulse and nuclear-thermal) show up here. I don't know if I have enough knowledge on all the relevant fields to write a good answer for you but if no one else offers it up there's plenty online. These seem like valid approaches that are mostly engineering (and maybe political opinion) challenges rather than requiring breakthroughs in physics/materials/etc. $\endgroup$
    – ben
    May 5, 2019 at 18:14

1 Answer 1


For such speeds you'd either have to create fantastically efficient rockets or do away with rocketry altogether. I am limiting this answer to technologies that are created with known physics, otherwise I'm just writing science fiction.

Since we are also assuming that you want to slow down and enter orbit within reasonable time, and we are assuming you want to be able to do this with an arbitrary solar system body, certain technologies are automatically out. For example, solar sails wouldn't work to slow you down at Pluto, and magnetic sails wouldn't work if you have no magnetosphere to thrust against.

I'm also limiting this to technologies that have been demonstrated. Therefore, fantastical things like antimatter rockets are off the table.

To me, it seems the best bet is a beamed propulsion system. Such a system would not use rocketry at all, but a laser beam. This would do away with any fuel requirements, since...there is no fuel. Such systems are being demonstrated and it has been shown in the lab that reflecting the laser beam increases thrust dramatically. One proposal for an interplanetary railway seems very promising, and the problems with its realization are quite accessible to today's engineering research.

A big avenue for research without even building one would be trajectory design and laser network design. There are two major issues with this propulsion technique.

First, the beam is straight and orbits are curved, plus gravity is nontrivial when you consider how well you need to align the beams over planetary distances. Some work has been done on the problem of designing trajectories and laser aiming protocols, but it has been limited to the two-body problem only (see [here])1.

Second, the idea of the pulsed-laser propulsion method is to pulse a laser off of a reflector. This reflector is, in theory, in orbit of something and therefore the pulses off of it will cause a momentum change. This momentum change needs to be cancelled. What is the most efficient way of doing this? Is it possible to create a network of these things that beam off of each other to cancel momentum with the "railway" is not in use? How does it depend on the planet or other body that the reflector is orbiting? If you use a rocket-based stationkeeping system, then you need to refuel it. How do you do that most efficiently?

Network design is an interesting problem as well. Logically, you wouldn't need a station at every body you want to go to, although you might want one. For instance, a network of these things spread across the asteroid belt might enable travel anywhere inside the asteroid belt, presuming that the beams do not degrade significantly over large distances (a HUGE if that requires a lot more power to reduce). What is the minimal network that enables exploration of the entire solar system up to some distance of interest?

Large-scale space power systems would be another interesting one.

Learn more here.

  • $\begingroup$ When I was reviewing this question yet more for polish, I wondered about the possibility of massive particle beams, as opposed to massless beams (my post briefly touched this and remarked that the power requirement is insane). The trick with beam propulsion is then we move from the "tyranny of the rocket equation" to the "asymptotically gentler" tyrant of the Einstein mass-energy-momentum relation. $\endgroup$ May 5, 2019 at 10:59
  • $\begingroup$ But it really seems like the separation in terms of technological difficulty, that some answers/comments to the referenced question hinted at, between "interplanetary" and "interstellar" travel does indeed blur when you want things to become economical in the first case. $\endgroup$ May 5, 2019 at 11:04
  • 1
    $\begingroup$ I also made sure to focus only on technologies that have already been demonstrated. There are lots of concept technologies that have been thought about, but none that anyone has built. Otherwise the answer is "antimatter rocket" $\endgroup$ May 5, 2019 at 11:04
  • $\begingroup$ I'm curious though, what do you mean by "trajectory design and network design"? What outstanding problems are there in those fields? $\endgroup$ May 5, 2019 at 11:05
  • 1
    $\begingroup$ @The_Sympathizer I've seen massive beam designs for interstellar use, but the same idea would work interplanetary. Your beam consists essentially of a series of small very tough guided missiles which can be launched electromagentically and steered a little. When they get close to the spaceship you blast them into plasma with a laser or electron beam (or maybe they blast themselves into plasma with a small nuke or something) You can then use a magnetic field to extract momentum from the plasma. (the same field and lasers could probably also provide some impact protection). $\endgroup$ May 5, 2019 at 12:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.