|Fig. 1: Shuttle Main Engine Test Firing. (Source: Wikimedia Commons. Courtesy of NASA)|
Mars is currently the primary arena for robotic, and potentially human, exploration in the solar system. Due to it's proximity to Earth, similarity in size, and the presence of water in different phases, Mars simultaneously represents a window to the past and a gateway to the future.  With the possibility that liquid oceans once existed on Mars, the ever present question is whether or not life ever existed or continues to exist on Mars; or, at a minimum, prebiotic chemistry.  With the resources of frozen water at the Martian pole, Mars also represents a candidate for future human colonization. 
These questions and opportunities drive the continued exploration of Mars, however to date, only a small fraction of the planet has seen direct exploration. Missions to Mars that are capable of exploring large areas, such as orbiters, have thus far been incapable of performing the direct, in-situ observations that are necessary for determining the potential for biochemistry; e.g. drilling and mass spectroscopy. On the other hand, robotic missions that have performed these detailed observations have been severely restricted in the area of the planet they have been able to explore. For example, the rover Opportunity - the longest travelled Martian rover - has covered a distance just exceeding 40 kilometers in the last 4000 days according to NASA. Furthermore, landers and rovers have been fundamentally limited in the broad areas they may land and explore due to the need for a large, unobstructed landing zone to fully contain the landing error elipse. This often precludes some of the most attractive regions for geologic exploration (citation).
The ideal Mars exploration vehicle would be one that can perform direct, in-situ observations and measurements. The vehicle would be capable of traversing a wide range of the Martian terrain and precisely maneuver itself into geologically interesting terrain. The vehicle should be able to perform exploration over a long time period, potentially even indefinitely such as what the Opportunity rover has achieved.
|Fig. 2: Description of the radioisotope thermal rocket thrust generation.|
A concept that satisfies these aims is the Mars Hopper powered and propelled by a radioisotope thermal rocket (RTR).  Like currently existing rover technology, the hopper is a mobile platform for planetary exploration. Unlike current rovers, the hopper derives its mobility from rocket propulsion instead of wheeled motion. The rocket propulsion allows the hopper to launch from one location on Mars and land in another - potentially many kilometers away - thus performing a "hop". By performing these low-velocity hops (low-velocity in comparison to the high-velocity atmospheric entry maneuvers that all landers must perform) the hopper would be capable of precisely navigating to narrow landing zones thus enabling the exploration of scientifically significant - yet difficult to access - locations. 
The essential subsystem of the hopper - the subsystem for which this paper focuses - is the propulsion system. At the heart of the propulsion system is a radioactive core composed of a material such as plutonium. The plutonium core is utilized in a radioisotope thermal rocket which is discussed in detail in the following section. The advantage to this system is that it can double as the electrical power subsystem and, potentially, triple as the thermal management subsystem, too.  The power subsystem could be a derivative of a commonly used radioisotope thermal generator (RTG). The excess decay heat of the radioactive core can be used for thermal regulation during the Martian night. Furthermore, the RTR concept may be able to employ in-situ resource utilization to collect propellant from Mars' atmosphere and avoid the need to carry all of its propellant to Mars. 
The fundamental method for thrust production in a RTR is the same as almost all chemical rocket engines: a hot gas is expanded through- and ejected from a supersonic, converging-diverging nozzle. What differentiates RTRs from other chemical rockets is the gas that is expanded and the heat source for heating the gas. In classical, chemical rockets heat is produced via combustion and the propellant that is expanded and ejected are the products of the combustion. An illustrative example would be the Space Shuttle Main Engines (SSME), see Fig 1. The SSME combusts liquid hydrogen and liquid oxygen (LOX) to produce heat; i.e. the thermal energy in the combustion products. This thermal energy is then converted to kinetic energy through a supersonic, converging-diverging nozzle. The ejection of these high-velocity combustion products from the nozzle is how thrust is generated. See Fig. 2 and 3 for illustrations
|Fig. 3: Basic diagram of radioisotope thermal rocket.|
For later comparison it is noted that the SSME is capable of a specific impulse of ~450 seconds in a vacuum (according to Aerojet Rocketdyne). Note that specific impulse is a measurement of the thrust produced with respect to the amount of propellant exhausted per unit time. While a more thorough discussion of specific impulse is omitted from this paper, it can be thought of as the primary measurement of efficiency of a rocket.
In a radioisotope thermal rocket no combustion occurs to produce heat. Instead a gas propellant is heated by the decay heat produced during radioactive decay of an element such as plutonium. A mass of plutonium contained in the rocket is allowed to decay and the decay heat is trapped in some form of thermal capacitor such as beryllium.  Once the thermal capacitor has reached a desired operating temperature, the propellant is blown around or through the thermal capacitor, thus heating the propellant and cooling the capacitor. The hot propellant is then exhausted through a converging-diverging nozzle, producing thrust.
Since combustion is not required there is no need to use combustable combinations and a single propellant can be selected to optimize efficiency or utilize available resources. At a given temperature hydrogen is the most efficient propellant; i.e. capable of producing the highest specific impulse. Models of an RTR system have shown that hydrogen could produce specific impulses on the order of 700 or 800 seconds. However there are issues with storing hydrogen as it is a cryogen and very low-density. For a Mars exploration mission, there may be a better alternative. Since combustion is not necessary, any gas could be used as a propellant. To this end it is proposed that Mars' own atmosphere be collected and used as the propellant (in-situ resource utilization).  This eliminates the need of carrying difficult-to-store propellants all the way to Mars. Implementing a atmospheric collection system that could extract carbon dioxide as the main propellant could lead to specific impulses on the order of 200 seconds. Combine this in-situ resource utilization with the reusable aspect of the plutonium heating element and you have a reusable rocket that could launch and land an indefinite number of times. A single hopper could potentially traverse the whole of Mars' surface.
© Ross Allen. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
 M. H. Carr, Water on Mars (Oxford University Press, 1996), p. 197.
 V. R. Baker, ete al., "Ancient Oceans, Ice Sheets and the Hydrological Cycle on Mars," Nature 352, 589 (1991).
 H. R. Williams, R. M. Ambrosi, and N. P. Bannister. "A Mars Hopping Vehicle Propelled by a Radioisotope Thermal Rocket: Thermofluid Design and Materials Selection," Proc. R. Soc. A 467, 1290 (2011).
 S. D. Howe , "The Mars Hopper: an Impulse-Driven, Long-Range, Long-Lived Mobile Platform Utilizing in situ Martian Resources," Proc. Inst. Mech. Eng., Part G, J. Aerosp. Eng. 225, 144 (2011).