Nuclear Propulsion A Historical Perspective

A by-product of the need for carrying the heavy atomic and thermonuclear bombs of the 1940s, nuclear propulsion was explored in great depth in the US and Soviet Union from the late 1940s throughout the 1950s and until the early 1990s. In the US the rationale for starting its development (by the Atomic Energy Commission, AEC, in 1953, through the program ROVER) was the perceived need for a 75,000 lbf thrust nuclear thermal rocket to power the third stage of US intercontinental ballistic missiles (ICBMs). In fact, in 1956 USAF joined ROVER, but after the Atlas ICBM was flight-tested in 1958, NASA with AEC (i.e., its Los Alamos Science Laboratories, LASL) were charged to replace USAF as the ROVER Program leaders. In 1961 this effort branched out (via contracts) to Westinghouse and Aerojet General; the industrial branch of ROVER was called NERVA (Nuclear Engine for Rocket Vehicle Applications).

The original organization chart of NERVA can be found in [Howe, 1985]. An entertaining history of ROVER/NERVA, focusing mainly on its US politics, can be found in [Dewar, 2004]; all technical work can be found in final report form [Westinghouse, 1972]; synopses can be found in [Bohl et al., 1989; Howe, 1985; Gunn, 2001; Rose, 2008]. An excellent summary of the technological path of ROVER/NERVA can be found in [Gunn and Ehresman, 2003].

Kiwi B-4

Kiwi B-4

cross section

Figure 7.13. Diagram of a NERVA Kiwi nuclear reactor showing a single fuel bar cross-section (see [Gunn, 2001, Fig. 2]).

cross section

Figure 7.13. Diagram of a NERVA Kiwi nuclear reactor showing a single fuel bar cross-section (see [Gunn, 2001, Fig. 2]).

The ultimate purpose of ROVER after 1958 was to develop reliable, safe, and efficient nuclear reactors for space applications. The first phase of ROVER progressed at the Los Alamos Science Laboratories (LASL) through a series of proof-of-principle Kiwi reactors (Kiwi-A, Kiwi-B), each with variants testing different fuel bars, geometry and materials. During this first phase, for instance, Kiwi-B4, an advanced design shown schematically in Figure 7.13, and on its test stand at Los Alamos in Figure 7.14, was tested at 1,030 MW. In 1961 program NERVA I started: its purpose was to engineer Kiwi reactors into rocket engine prototypes and to test them. NERVA spawned the NRX family of "engines" (six in all). For instance, NRX-A3 was derived from Kiwi-B4, and tested at 1,165 MW. The general scheme of all NRX rocket engines is that of Figure 7.18 (see p. 312).

The power of Kiwi reactors was about 1 GW, to support a projected rocket thrust of 50,000 lbf. In 1965 Kiwi designs started evolving at LASL into PHOEBUS 1 and 2. Evolution was based on fuel rod technology and reactor diameter, that went from the 35 in of Kiwi B4E to the 55 in of PHOEBUS 2 with a commensurate power increase. On the industry side, these reactors were considered the precursor of the second phase of NERVA (NERVA II). The PHOEBUS family of reactors was the most powerful ever (see Figure 7.15, showing PHOEBUS 2 on its test stand at Los Alamos).

Figure 7.14. The NERVA Kiwi B4-E reactor on its test stand at Los Alamos [Dewar, 2004].

Meanwhile industry was concentrating on rocket engine lifetime. By that time NP was considered essential to manned Mars missions. The NRX-A5 and 6 engines developed during the NERVA I phase were tested at more than 1 GW for up to 62min. At the time all space missions planned assumed the engine needed to work for no more than 1 or 2 hours at most, but also to be capable of multiple restarts. A schematic drawing representative of the NRX family of engines built by Westing-house is shown in Figure 7.16. The NRX family was based on the Kiwi B4E, that had fuel no longer in the form of uranium oxide but in the much more heat- and corrosion-resistant uranium carbide. Tests were in fact performed at a steady 2,200 K reactor temperature.

PHOEBUS progressed at LASL through versions -1A, -1B, and culminated in -2A, that reached 4,082 MW for 12.5 min. Right at that point PHOEBUS funding was suspended, mostly because the engine that could be derived did no longer have a specific mission. However, work was continued on much smaller research reactors (PEEWEE, 500 MW power) that were less time-consuming and less expensive to build, test and operate, focusing on improving fuel rods durability while raising temperature and reliability, with industry following suit in the parallel NERVA program.

At the end of the program in 1972, the NERVA NRX ETS-1, the last nuclear rocket engine developed, was tested at 1,100 MW for a total of 3 hours 48 min. ETS-1 was conceived as the direct precursor of the final NERVA I engine shown in Figure 7.20. The nominal power planned for the final NERVA I rocket was 1,500 MW, with Isp = 825 s. By design, this engine was capable of 10 restarts

Figure 7.15. The 4 GW PHOEBUS 2 nuclear reactor on its test stand at Los Alamos [Dewar, 2004].

lasting 1 hour each. Its reliability was projected to be 0.997, that is, more than 10 times better than any current LRE. The weight was estimated at 15,000 lb, the thrust 3.34 x 105N. Power density was «2MW/dm3 (200 times greater than in gasoline engines). In short-duration tests, bursts of power reached 2 x 105 MW and thrust 8.9 x 105 N [Lawrence et al., 1995]. Future upgrades were planned assuming Isp up to 900 s, since progress in high-temperature materials was supposed to enable reactor operation at 2600 K.

Nozzle skirt extension

Internal shield

Control drum

Nozzle skirt extension

Internal shield

Control drum

Propellant line

Figure 7.16. Schematic diagram of the Westinghouse NRX nuclear engine [Dewar, 2004].

External Reflector disc shield

Propellant line

Figure 7.16. Schematic diagram of the Westinghouse NRX nuclear engine [Dewar, 2004].

External Reflector disc shield

Still in 1972, LASL did a definition study of a 16,000 lbf thrust NTR weighing 5,890 lb (including the shield) that could be carried to LEO by the US Shuttle, at that time in the planning stage. This nuclear engine was proposed to power interplanetary missions, but also to drive a ''space tug'' from LEO to GEO and other orbits [Gunn and Ehresman, 2003]. However, because of cost, declining political support, lack of a clearly defined mission, and other reasons, this program came to an end during the Nixon presidency. The many lessons learned during the tests carried on at Los Alamos for the ROVER program are summarized in a Los Alamos report [Koenig, 1986]; an extended account of the ROVER/NERVA programs is available in the Encyclopedia of Physical Sciences and Technology [Meyers, 2001].

USAF kept working in nuclear propulsion under the SNTP program until 1993, with an annual budget of about $40 million. Much of this work was spent in finding ways to make space nuclear reactors more compact and capable of standing higher operation temperature and/or more power cycles, and resulted in the Particle Bed Reactor (PBR) and CERMET concepts briefly described in Sections 7.9 and 7.10. Sponsored by USAF, classified work in NP using PBR starter in 1983 in the context of the Space Defence Initiative (SDI) of President Reagan, dubbed ''Star Wars'' by the media. By 1987 the classified project name was Timberwind [Rose, 2008]. Its purpose was to design NP systems to lift nuclear directed-energy weapons (X-ray lasers) to orbit. To improve performance, the propellant was slush-H2, 16% denser than liquid hydrogen at a solid/liquid fraction of 0.5 [Ohira, 2004]. The LiH moderator and PBR topology resulted in an engine about half the weight and volume of the last NTR developed under ROVER. The ramp-up and ramp-down time was also much reduced, of order 10 s. By 1990 three Timberwind engines (so-called -45, -75 and -250) were under development, but in 1992 this program was allegedly terminated. Because Timberwind is still classified, no more details are available.

Not much is left right now of NERVA. A mock-up of its final ETS engine is standing in the NASA Space Park in Huntsville, Alabama (see Figure 7.17). Conceptual work in NTR is still being carried on at NASA-Glenn Research Center by a team led by Stanley K. Borowski, who keeps in touch with the ''old-timers''. This team studies and updates continuously this technology in view of a future manned Mars mission. In its latest architecture the concept has evolved into a NTR working both as a propulsion system and as a modest power generator (^110 kW). This is the so-called ''bimodal propulsion concept''. The power generator is supposed to be used for instrumentation, support of crew activities, data transmission and refrigeration of the liquid hydrogen propellant during the trans-Mars and trans-Earth (return) mission stages [Borowski et al., 2000]. In this context, a mission to the asteroid belt has been proposed as a ''dry run'' prior to any manned Mars flight. This shows a cautious, albeit perhaps too costly, approach, if Mars had been selected as primary mission objective under the 2004 Space Exploration Initiative of President G.W. Bush.

The Russian work, until recently shaded in secrecy, is now better known (e.g., see the work in Goldin et al. [1991], Rachuk et al. [1996], Ponomarev-Stepnoy et al. [1999], Demyanko et al. [2001], Konyukov et al. [2004], Dewar [2004], Koroteev et al.

Figure 7.17. Mock-up of the NERVA 1 as it stands in Huntsville, Alabama, Space Park [Dewar, 2004].
Figure 7.18. Simple scheme of a nuclear thermal rocket fed with liquid hydrogen.

[2007]). Some of the most important steps in the development of NP propulsion in the former SSSR are reported in [Rose, 2008]. The origins go back to Prof. M.K. Tikhonravov, at the Soviet Academy of Sciences, while working with Korolev's OKB-1 in the early 1950s. For a manned Mars mission Tikhonravov's calculations indicated that such a mission would require lifting some 1,600 tons to LEO. Even with the N-1 launcher designed for the Soviet Union manned Moon shot, the sheer number of launches, and their cost, was staggering. Hence the interest in NP. At the same time the military were looking, just as in the US, at propulsion for their ICBMs: calculations in the early 1950s had predicted that a NP-powered singlestage missile could reach any target on Earth. It was at that time that Keldysh (the head of the Soviet Academy of Sciences), Korolev (at OKB-1) and Kurchatov (the head of the Soviet nuclear organization) started their collaboration and became known as "the three Ks''. Two ICBM designs by two different bureaus followed in the period 1958-1959, one by Glushko and the other by Bondaryuk. Glushko's used NH3 as propellant (instead of the LH2 of ROVER) since it posed fewer logistic problems than liquid hydrogen, and his engine had a planned thrust of 1,255 kN. Bondaryuk chose a mixture of NH3 and methyl alcohol.

Just as in the US, rapid progress in LRE convinced the military that NP was no longer necessary to their ICBM. Nevertheless, investing in NP continued, and by 1961 two LH2-fed NTR engines were designed, one for an upper stage, with thrust of order 30-40 tons, and a second, more powerful engine called the RD-600. These two engines continued to be developed in the 1962-1970 timeframe. In 1971 all work on NP was assigned to NPO-Luch, in Podolsk, Russia, a company specializing in treating nuclear fuels and high-temperature materials. NPO-Luch continued to develop NP for the next 18 years. Their Baikal-1 NTR was designed, assembled and bench-tested at least 30 times, proving that the overall architecture of the engine was very reliable. Although details remain sketchy, it is known that two more NTR engines were designed and assembled, the small RD-0410 (thrust 3.5 tons) and the RD-0411 (thrust 70 tons). Both engines were neutron-driven, and were extensively tested in "cold" (no fission) and "hot" (fissioning) mode in the secret Semipalatinsk-21 facility. However, after the Academy of Sciences rejected an overly ambitious manned Mars mission project by the Chelomei OKB (the Mars spacecraft alone weighed 1,100 tons), all development work for Mars missions stopped in 1972, although, as said, engine development continued until the collapse of the Soviet Union. It is important to note that, as in the US, just when the technology had matured and demonstrated its outstanding performance over that of LRE, it was a political and economic decision—not engineering issues— that killed all work on NP hardware.

Russian sources claim to still retain technology and especially testing capabilities. In fact, the Russian NiiCHM organization has developed very high temperature materials (>3,000 K) that would be invaluable in building a future highperformance NTR. According to Gafarov et al. [2004], nuclear reactors in the 1,000-kW range are still being investigated or tested for interplanetary missions both for thermal rockets and for power generators for NEP, and a conference on nuclear propulsion has taken place in Moscow in May 2005. This conference was the last in a series organized by the NIKIET Agency and was sponsored by the Russian International Science and Technology Center (ISTC), which also works as a clearinghouse for information in NP [Pradas-Poveda, 2008]. In fact, the dual-mode NTR projects reported in [Koroteev et al., 2007] include a nuclear space engine concept in the 340 MW class, designed to work at 2,900 K using a U/Zr carbide/nitride fuel. At a stagnation pressure of 60 bar, the design vacuum thrust is 68 kN with Isp close to 960 s, for a total engine mass (including shielding) of 12.2 ton. The level of detail of this project indicates that interest in Russia for NTR technology is still high, and that work at the study level keeps being funded.

To conclude, perhaps the most significant aspect of this short survey is that it shows that NP is not a new topic. Work in the 1960s and 1970s at LASL produced NTR reactors capable of 4 GW power. At a conservative Isp = 800 s this figure would ideally produce thrust of order 5 x 105N (50 tons). Reactors and engines were designed and built with technology and especially design practices that many now would brand obsolete, because it used computers an order of magnitude slower than now available. Performance figures and achievements should give pause to people having second thoughts or misgivings when discussing 25- or 50-kW NEP thrusters for the JIMO missions planned by NASA and described later.

Was this article helpful?

0 0
Solar Panel Basics

Solar Panel Basics

Global warming is a huge problem which will significantly affect every country in the world. Many people all over the world are trying to do whatever they can to help combat the effects of global warming. One of the ways that people can fight global warming is to reduce their dependence on non-renewable energy sources like oil and petroleum based products.

Get My Free Ebook


Post a comment