|Fig. 1: The Stirling cycle depicted on a pressure-volume graph. (Source: Wikimedia Commons)|
Among the many challenges of spaceflight, long-endurance energy sources stand out as particularly difficult. Solar power arrays are a common solution for satellites, but for a vehicle blocked from the sun by atmospheric or other interference, or one whose mission takes it sufficiently far from the sun, they are not a viable option. High-capacity batteries might serve for low-wattage applications, but the weight problems posed by carrying an energy source with relatively low energy density per unit mass have yet to be overcome. The hazards of launch obviate a number of other potential solutions that are too vulnerable to the G-forces or vibration loads incurred while reaching orbit.
In numerous cases, particularly for missions whose flight profiles require continuous operation for long periods in deep space, space programs have turned to the radioisotope thermoelectric generator. The design and operation of such devices has been thoroughly covered in previous reports. [1-4] A basic synopsis of such systems is provided below.
Decay of radioactive elements releases, among other things, a nontrivial amount of heat, which can serve as an energy source for a mission. The amount of heat released is proportional to the amount of the radioactive core remaining, so isotope selection must be informed by the intended mission duration. Common selections include isotopes of Plutonium, Curium, and Strontium; however, the considerable diversity of radioactive isotopes in existence offers a number of nontraditional options. Radioisotope thermoelectric generators have been used for decades, from the Apollo missions to the Mars rovers and the New Horizons probe. 
The basics of harvesting energy from a thermal gradient are well understood; they are used routinely on Earth in geothermal power or thermal plants fueled by coal, natural gas, or nuclear power. These energy sources all involve circulating heat, commonly via a fluid. Space agencies have long used a less complex version of this process, involving thermocouples that convert a heat gradient directly into electricity. This design sacrifices energy efficiency for mechanical simplicity; since maintenance is impossible on a mission of duration and distance sufficient to justify such a power source, it is best to eliminate moving parts wherever possible.
Standard thermoelectric generators of the variety used in space are roughly 8% efficient.  This represents a major design constraint. Energy source efficiency is very important. Not only is weight at a premium, given the energy requirements for boosting mass into orbit, but minimizing the amount of radioactive material necessary for a given power output reduces costs and decreases the hazards that would result from a catastrophic launch failure.  Additionally, until very recently, Pu-238, NASA's isotope of choice, was not being produced at scale. Stocks came from now-depleted Cold War-era reserves and from Russian sources.  Lead times are long and the fuel is hazardous and expensive; the incentive is clearly present to create a more fuel-efficient system.
Enter the Advanced Stirling Radioisotope Generator (ASRC), which offers a per-kilogram fuel efficiency four times greater than its thermoelectric competitor.  The Stirling cycle is complex, but it does have similarities to other thermodyamic cycles. Like the Carnot cycle, it consists of four steps that result in delivery of net work. One principal difference is that the isentropic compressions and expansions of the Carnot cycle are replaced by two regeneration processes that occur at constant volume. Additionally, the phases of the Stirling cycle are not discrete, which gives rise to the great complexity of the cycle. A Stirling cycle illustrated on a conventional pressure-volume diagram (see Fig. 1) looks like an ellipse, not a quadrilateral.
The critical difference between the ASRC and prior generations of radioisotope thermoelectric generators is that it uses an oscillating piston to manage heat flow to and from the hot and a cold reservoirs.  As mentioned above, space agencies avoid mechanical complexity and maintenance risk wherever possible; however, the fourfold increase in efficiency was deemed worthy of investigation. The oscillating piston is magnetized, and its movement through a solenoid induces a current, which translates to power for the spacecraft. This movement is so rapid -- about 100Hz -- that it properly induces an alternating current to feed the spacecraft electronics.  In turn, this places stringent requirements on reliability for the system. The Voyager 1 mission launched 40 years ago, and it still has power. If Voyager 1 contained an ASRC, its piston would have oscillated about 125 billion times.
The Stirling cycle has been known, and engines have been designed to use it, since the early 1800's. However, the ASRC represents the first application of the technology to space. A joint contract was issued to develop the device in 2000; the Department of Energy and NASA Glenn engaged Lockheed Martin Space Systems to develop and produce the device.  The project met a number of critical design steps; multiple functional prototypes were produced, and as of 2009, a hermetically sealed pair of devices had successfully completed 16,000 hours of testing.  Unfortunately, development of long-endurance systems is by nature a very lengthy process. Completing 16,000 hours of testing takes two years if the device is in continuous operation.
Given these constraints, as well as trends in aerospace procurement, it should perhaps not come as a surprise that the ASRC program ran massively over-budget. The program was expected to cost $150 million, but by 2013, it had run over-budget by $110 million with no end in sight. NASA cancelled the progarm, taking the produced hardware and design and using the savings to support Pu-238 production; given that missions will be forced to rely on less efficient systems, requiring more Pu-238, for a greater time than expected, then this is a sensbile move.  Development of the ASRC will continue, albeit at a slower pace, and NASA now expects to deploy the device on missions launching in 2028.
The ASRC program offered a power source that greatly increased mission flexibility. A mission mounting an ASRC, relative to those using earlier generations of radioisotope generators, could mount more hardware, endure longer, present fewer launch hazards, and cost significantly less. Even at the cost of significant cost overruns, this is a goal worth pursuing; NASA rightly considered the research an investment in the diversity of future missions profiles. As people who benefit from the exploration and development of space, we should hope that NASA manages to continue ASRC development and testing in a more rapid, cost-efficient manner than its industry partner.
© Evan Long. The author warrants that the work is the author's own and that Stanford University provided no input other than typesetting and referencing guidelines. 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.
 J. Ruffio, "What Future for Radioisotope Thermoelectric Generators (RTG)?" Physics 241, Stanford University, Winter 2017.
 A. Crerend, "Radioisotope Thermoelectric Generators (RTGs)," Physics 241, Stanford University, Winter 2015.
 M. Jiang, "An Overview of Radioisotope Thermoelectric Generators," Physics 241, Stanford University, Winter 2013.
 V. Chirayath, "Advances in Thermophotovoltaic Radioisotope Generators for Deep Space Exploration," Physics 241, Stanford University, Winter 2015.
 "Advanced Stirling Radioisotope Generator," U.S. National Aeronautics and Space Administration, NF-2013-07-586-HQ, July 2013.
 P. C. Schmitz, L. W. Mason, and N. A. Schifer, "Modular Stirling Radioisotope Generator," U.S. National Aeronautics and Space Administration, NASA/TM-2016-218911, April 2016.
 W. A. Wong, S. D. Wilson, and J. Colline, "Advanced Stirling Convertor Development for NASA Radioisotope Power Systems," U.S. National Aeronautics and Space Administration, NASA/TM-2015-218461, April 2015.