STM-D-1047Paper2004Published and peer-reviewed

Nuclear systems for Mars exploration

Tibor S. Balint

Summary and citation · read the original at the source · none found

In one page

Tibor Balint, at NASA’s Jet Propulsion Laboratory, does the arithmetic nobody enjoys and everybody needs: how much power can you actually get to Mars, and what stops you. He calls the answers breakpoints — the limits past which today’s technology simply cannot go and something new is required. His method is deliberately blunt. Reduce every mission to three quantities, mass, power and time; split it into three stages, the trip out, the time in orbit and the time on the surface; and take the bounding case for each. The result is a ledger. Rockets cap what you can launch, so they cap the reactor you can fly; parachutes and heat shields cap what you can land, so they cap the reactor you can put on the ground. Balint’s own framing is the interesting part: he calls this the mass-dependent paradigm, and sets against it a future paradigm in which power is no longer rationed by what a launch vehicle can lift.

Why it matters hereChapter 8 asks what propulsion looks like when it stops being about throwing mass overboard, and this paper is the honest ledger of the architecture that still does — the one every vacuum-powered design is measured against; chapter 12 is about the energy substrate, and Balint shows precisely how much of a Mars programme is decided by how many kilowatts you can get to the surface.

What it claims

01The assessment architecture is three simplifications used together: a lumped-parameter approach that reduces every mission to mass, power and time and derives the rest; a bounding-case approach that looks only at upper limits, so that technology breakpoints fall out; and a decomposition of a generic Mars mission into a transportation stage, an in-orbit stage and an on-surface stage, each characterised by those same three quantities.Section 2, Assessment Architecture
Published and peer-reviewed
02In-space propulsion and power generation are sized by what the launch vehicle can deliver and by the trajectory chosen, not by the reactor. On a direct approach to low Mars orbit the total mass is limited to about 8 tonnes; to low Earth orbit it is about 20 to 24 tonnes. A 100-kilowatt nuclear-electric system could deliver about 15 tonnes to Mars in about 1400 days, most of that time spent spiralling out of and into the two gravity wells, while a nuclear-thermal system could deliver about 13 tonnes in about 200 to 350 days depending on trajectory.Section 3, Transportation Issues, and Section 5, in-space power generation
Published and peer-reviewed
03Nuclear thermal propulsion divides into three reactor categories with specific impulses of 800 to 1100 seconds for a solid core, 3000 seconds for a conceptual liquid core and 6000 seconds for a conceptual gas core. For the same initial-to-final mass ratio a nuclear system achieves twice the velocity change of a chemical one, which would cut the one-way trip to Mars from about six months to about two or three, or alternatively double the delivered mass for the same trip time; nuclear thermal and nuclear electric systems may reach velocity changes of 22 to 33 kilometres per second.Section 3, Transportation Issues, Table 1 and the nuclear thermal propulsion discussion
Designed, not yet built
04Surface power is capped by entry, descent and landing rather than by reactor design. With a Viking-type aeroshell and parachute the maximum landing mass on Mars is about 2 tonnes at low elevation in good weather and about 1 tonne at high elevation in adverse weather, supporting about 15 to 20 and about 5 to 7 kilowatts of electrical power respectively; the same aeroshell with a Mach 3 parachute raises the limit to about 3.6 tonnes and about 25 kilowatts. Landing accuracy with entry guidance and optical navigation is about 3 kilometres, which by itself rules out multiple landings that share power.Section 4, EDL Issues, and Section 6, Conclusions
Published and peer-reviewed
05The headline breakpoints for the technology of the day: in-space nuclear fission reactors can provide about 100 to 200 kilowatts of electrical power for orbital missions at low Mars orbit and about 10 to 25 kilowatts for surface missions. Small surface reactors become more mass-efficient than solar panels somewhere around 3 to 5.5 kilowatts, and batteries and fuel cells are ruled out for anything measured in months or years, leaving reactors and radioisotope generators. Power beaming would need two orders of magnitude of improvement in conversion efficiency, from about 0.4 per cent to about 40 per cent, before it could be seriously considered.Section 5, Power Generation Issues, and Section 6, Conclusions
Published and peer-reviewed
06What to watch is the paradigm Balint names but does not price. Crewed Mars bases will require hundreds of kilowatts or even multiple megawatts and several hundred tonnes of landed mass, which no line in his ledger reaches; he describes today’s space exploration as characterised by mass dependence and consequent power limitation, and sets against it a future paradigm in which advanced propulsion and power sources provide power far beyond current limits, naming fusion and antimatter as the distant-future candidates and fission as the near-term one.Section 1, Introduction, Figure 2, and Section 6, Conclusions, closing paragraphs
What to watch

The way in

https://doi.org/10.1109/aero.2004.1368102SOURCE READ IN FULL, BUT NOT REPRODUCIBLE — AND WHY NOT. The complete twenty-page paper was retrieved and read on 2026-09-08 from the JPL Open Repository, handle 2014/8042, file 03-3452.pdf, reached through the JPL Dataverse search interface that trs.jpl.nasa.gov now redirects to. It was checked for public-domain status and it does not qualify. The first page carries the publisher’s own copyright notice, 0-7803-8155-6/04/$17.00 followed by the copyright symbol and 2004 IEEE, together with the line IEEEAC paper 1118, version 3, updated 19 November 2003. The NASA Technical Reports Server holds the same work as citation 20210001792 and its copyright block is explicit: belongs-to-publisher is true, belongs-to-US-Government is false, and the approver’s note reads that items in the Technical Report Server are protected by copyright but are furnished with United States Government purpose use rights. The record carries no downloadable file there. The JPL Dataverse deposit is tagged CC0 1.0, but that tag sits on the repository’s dataset wrapper and not in the document, which carries the publisher’s notice instead, so it is not treated here as a licence on the paper. This page therefore reproduces no text of the paper at all, not even the abstract, and every locator below points to a numbered section, table or figure of the document that was read. The work was performed at the Jet Propulsion Laboratory, California Institute of Technology, under contract to NASA; the author’s address as printed is 4800 Oak Grove Drive, M/S 301-170U, Pasadena, California. REGISTRY CORRECTIONS. The fetched record carried no year and gave the author as T.S. Balint with the title running into two trailing footnote markers; the paper prints Tibor S. Balint and the conference is the 2004 IEEE Aerospace Conference, so the year is 2004. NUMBERS. The paper writes MT for metric tonnes and kWe for kilowatts of electrical power; both are spelled out below. Its own figures are approximate and it says so — every value it quotes is prefixed with a tilde in the original.

How to cite it

Tibor S. Balint (2004) Nuclear systems for Mars exploration. doi:10.1109/aero.2004.1368102

Where it sits in the curriculum

Inertial mass reduction, the Navy patents and transmedium craft Fusion machines: pinches, focus devices and inertial drivers