Saturday, July 14, 2012

The Atomic Subterrene

Those Magnificent Men and Their Atomic Machines

The Atomic Subterrene

The Atomic Subterrene is a very atompunkish name. It sounds like a gadget Tom Swift might invent, and which would then be stolen by vaguely Slavic communists. It doesn't help that, if you google it, you'll find several hundred webpages of nonsense using pictures of the New York subway to explain how the elites are building secret underground clubhouses to ride out 2012.

But, for all the silliness that seems to attach to the name, the atomic subterrene was a very real, very serious idea, developed by Los Alamos Scientific Laboratory (LASL) in the 1970s. It was a startlingly simple proposition: rather than drill through rock, the atomic subterrene would use heat from a nuclear reactor to melt through it, digging wider tunnels faster and more efficiently than a conventional Tunnel-Boring Machine.

The Beginning
In the 1950s, atomic rockets were all the rage. The Atomic Energy Commission, the Air Force, and later NASA were running a program called Project ROVER, to develop a series of reactors for rocket propulsion. Meanwhile the Air Force's Aircraft Nuclear Propulsion Office was working on nuclear-powered turbojets and ramjets, such as the infamous PLUTO.

These propulsion reactors shared two common traits. First, because the efficiency of a propulsion reactor is determined by its temperature, the ROVER and ANPO reactors were designed to run at much higher temperatures than conventional reactors for ship propulsion or electricity generation. And second, since they needed to fit on planes or spaceships, they were designed to be very small, both in mass and volume.

One of these reactors was designed for Project DUMBO by the CMF-4 group at LASL. The DUMBO reactor consisted of a honeycomb of tungsten and uranium, through which hydrogen gas would be pumped; the nuclear reaction would heat the gas and thus produce thrust. To test the concept, LASL built a mockup using an electrical heating source in place of the uranium. Gas pumped through the mockup reached 3000o C, an impressive demonstration of the concept.

But, in 1959, DUMBO development was cancelled in favor of the graphite reactor designs that ultimately became Project NERVA. The CMF-4 group was reassigned, told to spend six months doing brainstorming on anything except rocketry.

The group explored all manner of exotic ideas, but only one of them concerns us today. During this period, one of the project members, Bob Potter, reread Edgar Rice Burroughs' novel At the Earth's Core, and started to think about ways to get through rock more efficiently than by grinding it up. He considered the idea of simply melting the rock out of the way – and he thought of the high-temperature tungsten heating elements used in testing the DUMBO concept.

Potter borrowed a few pieces of local basalt stone from a nearby highway construction site, and the group rigged up a tungsten heating element in the lab. Pressing the white-hot tungsten against the basalt quickly produced a neat little hole. Interestingly, the molten basalt flowed around the tungsten heating element, forming a sticky surface layer that shielded it from damage by air or water in the rock.

Figure 1: Tabletop Thermal Penetrator

Further experiments soon followed, culminating in a tabletop device with an outside diameter of 2 inches. Heated and pressed against rock, the penetrator would melt its way through. The molten rock would flow through a hole in the center of the head and out the back, where high-pressure gas would blow it to the surface. At this point the device was just called a rock-melting penetrator; the name subterrene had not yet been attached to it. The idea of using nuclear energy was not yet in the mix; the plan was for the penetrator to be powered by a connection to a generator on the surface.

Unfortunately, all good things must come to an end, and CMF-4's work on the penetrator was one of them. Bob Fowler, the head of the CMF group at Los Alamos, did not approve of the penetrator project, which he felt was not “proper” research. In 1962 he ordered the group to write up their results and move on to new projects, which they obediently did. The rock-melting penetrator was set aside, although that did not keep the Atomic Energy Commission from obtaining a patent on the idea.

Los Alamos Gets Back in the Game
The concept was not revived until eight years later. The Los Alamos staff had a habit of meeting at a local pub after work on Fridays to bullshit and kick ideas around in a more congenial environment. On one Friday, someone brought up the old rock-melting penetrator idea, and suggested upgrading it with more modern materials. It was suggested the concept could be improved by using heat pipes to connect a compact nuclear reactor to the tungsten heating element. Instead of heating the melting head with electricity from the surface, molten lithium would be heated by the reactor and pumped through the melting head.

The matter might have ended there if Los Alamos' congressional representative, Manuel Lujan, Jr., had not happened to wonder into the pub. When he stopped by CMF-4's table to ask what they were talking about, Eugene “Robbie” Robinson told him of their idea for a nuclear-powered tunneling machine.

Unfortunately, or perhaps fortunately, Representative Lujan misunderstood him and thought he was talking about an official Los Alamos program rather than a napkin-back discussion among off-duty scientists. He expressed his approval of the idea, and of the wisdom of the Atomic Energy Commission for sponsoring such a far-sighted, innovative project, and said he would contact the Joint Committee on Atomic Energy in Washington to express his pleasure with the program. This might prove rather awkward, since of course the AEC would have no idea what he was talking about.

Thinking quickly, Robbie phoned Norris Bradbury, the director of Los Alamos. Fortunately, Mr. Bradbury had a sense of humor about the whole thing – and, not only that, but felt the atomic penetrator was actually rather a good idea, and that the lab should organize a study of the concept!

During the spring, summer, and fall of 1970 a study group met to discuss the feasibility of the system. It was around this time that the device was given its name: the subterrene, as a terrestrial analogue to the submarine. In November the committee issued its report, “A Proposal for LASL Development of a Nuclear Subterrene,” recommending the paper study be expanded to a feasibility study, with the ultimate objective of building a device “capable of penetrating the earth to depths of tens of kilometers... To extend geological and geophysical exploration into the earth's mantle.” It was thought a subterrene capable of reaching the mantle could be built within 10 to 15 years.

This was not the first time someone had tried to apply atomic energy to tunneling. William Adams of Lawrence Radiation Laboratory had proposed building a “needle reactor” as a probe to the Earth's mantle in the early 60s, but the idea had gone no further than an article in Time magazine. Other machines for using atomic energy for mining had been patented, but enjoyed even less success.

The Los Alamos subterrene proposal, on the other hand, was a serious investigation backed by a major government laboratory. And the time was right for a radically new approach to drilling technology. The country's energy situation was deteriorating, and the AEC had been directed to look into development of non-nuclear energy sources in addition to its old mission of atomic energy. The subterrene could offer a number of new capabilities.

For example, conventional drilling can only produce circular tunnels, since the drill works by rotating, while the subterrene's melting penetrator head could be of any shape desired. The subterrene would have less environmental impact since it would produce little to no dust or vibration. It would last longer than conventional drills, which are quickly worn away by drilling through hard rock. It would require fewer personnel to operate.

But, most importantly, it was thought the system might be cheaper – the initial analysis suggested savings of up to $850 million dollars (1970 dollars) through 1990, on a development cost of $100 million.

Rowley and the other scientists speculated on a whole host of applications that might be opened up if the subterrene lived up to its promise. Aside from mining, excavating underground roads and pipes was an obvious use. Chemicals and gasses could be stored in underground chambers. Electrical energy could be stored in the form of underground pressurized air “batteries,” compressed in during periods of excess production and used to drive turbines when more energy was needed. The subterrene could dig storage cavities for toxic and nuclear waste, too deep for them to ever trouble the surface. The heat and pressure found deep underground could be exploited for chemical processing. Cities, even farms, could be extended underground.

The most promising application, however, seemed to be geothermal energy. The subterrene could be used to dig deep into the Earth's crust, to where the rock is heated by the mantle. Unlike a conventional drilling machine, since the subterrene worked by melting its way through, its efficiency would actually improve with depth. Two vertical tunnels would be drilled, side by side, and a great chamber or cavern excavated at the bottom connecting the two. Water would be pumped down one tunnel, be vaporized by the hot temperatures at the cavern bottom, and steam would then be pumped up through the other tunnel, where it would be used to drive a turbine. Ordinarily, geothermal energy can only be tapped in areas where near-surface hot rocks and groundwater coincide, but this sort of plant, called a Hot Dry Rock plant, would only require the rocks.

In December, the subterrene proposal was reviewed by senior personnel at LASL, with highly mixed results. One reviewer called it one of the dumbest ideas in history. The dominant view, however, was more favorable.

In April of 1971, the program was submitted to the National Science Foundation. Funding was ultimately approved through the Research Applied to National Needs program and work began in 1972, with the first patent for a nuclear or electrical melting penetrator filed the same year.

Rapid Excavation by Rock Melting
The program was aimed at developing both electrical and nuclear subterrenes. In the first year of the project, the scientists, under the able leadership of John Rowley, focused on developing small-scale prototypes powered by external electrical sources. These would serve as a proof-of-concept of the rock-melting drill, and would be useful and desirable in their own right.

In particular, the group was interested in a device they called a Geoprospector, which would be relatively small – about a foot in diameter – and would be used to retrieve samples of underground mineral deposits. Aside from the Geoprospector, proposed near-future applications focused on other relatively shallow, small-diameter projects, such as drilling gas pipelines or drainage tunnels.

Several prototype drilling machines were built in the first year, ranging in size up to 11.4 cm in diameter. One was tested at Bandelier National Monument near Los Alamos in May of 1973, digging drainage holes in Indian ruins.

Figure 2: Excavation at Bandelier National Monument

Unlike a mechanical drill, the melting penetrator didn't produce dust or vibrations that might damage the ruins. Five holes were drilled by the Los Alamos team, and then three more by National Park Service personnel. The NPS personnel reportedly found the machine easy to use, and were able to do the job with only minimal supervision. The drilling operation attracted a crowd of curious tourists, and it was decided to repeat the demonstration at a more politically useful venue – Washington, D. C.

The melting penetrator made its visit in mid-October. Four demonstrations were held at the Army Engineering Proving Grounds in Virginia over the course of two days in front of a crowd that included congressional representative, news media, and construction firms. A 50 mm penetrator drilled through a foot-thick slab of alluvium encased in steel, and then did it again in case anyone missed something the first time. Then, a second penetrator dug a vertical shaft. All told, about 300 people watched the demonstrations. A few months later, a third demonstration was organized in Denver, Colorado, and then in Tacoma, Washington.

Figure 3: Penetrator Demonstration at Washington, D.C.

Figure 4: Tunnel Produced by Thermal Penetrator

The Nuclear Subterrene
Although the small-diameter electrical prototypes were the focus of the team's efforts in the first year, design work continued on the nuclear system. Initial designs focused on building a nuclear-powered version of the melting penetrator, in which the reactor heat would simply melt through the rock. However, it soon became clear that it would be more efficient to use a combination of either rock melting and mechanical cutting, or rock melting and thermal fracturing. Patents on each of these concepts were filed on each in January of 1974.

Figure 5: Nuclear Subterrene for Soil and Soft Rock

The version pictured above was intended for use in soil and soft rock. Molten lithium would be pumped through a small nuclear reactor and circulated through the “annular melting penetrator,” which would reach temperatures of about 1,570o Kelvin, melting through a ring of rock in front of the machine. While the rock in the middle of the ring would not melt, it would be detached from the rock around and behind it, allowing it to be easily broken up by the rotating mechanical cutters in the middle of the machine's face. The rock that was molten would flow out and along the sides of the machine, being cooled by a heat exchanger to form a glass lining for the tunnel. In this design, the reactor heat is only a supplement to the cutters, which are analogous to the operation of a conventional tunnel-boring machine.

Figure 6: Nuclear Subterrene for Hard Rock

The second version was designed for harder rock. As in the first version, an annular melting penetrator would melt a ring of rock in front of the machine, detaching the material in the center from the rock behind it. Then, needle-shaped “fracturing penetrators” would use heat to break up the the core. The rapid heating of the rock would cause it to expand, but this expansion would be uneven, with some hot areas expanding quicker than other, cooler areas. The difference in expansion would cause the rock to fracture and break up.

As in the first version, the molten rock would be pressed against the side of the tunnel and cooled, forming a glass lining. The fractured rock would fall into a removal port, and be transferred to the surface for disposal by conveyor belt. Not visible on the diagram is a “clearing plate” that would periodically push forward from the fracturing penetrators, dislodging any stuck rock. Alternatively, the fracturing penetrators could be withdrawn by hydraulics to allow rock to fall past.

Safety was unlikely to be a serious concern. If anything went wrong, the reactor would be entombed underground. Even if leakage did occur for some reason – such as an accident in transport – the contemplated reactor designs would necessarily be very small, and therefore contain comparatively little radioactive material.

It was estimated that machines of these types could drill a 7.3-meter diameter tunnel at a rate of 1.5 meters per hour using a 25 MWth nuclear reactor. Hole diameters could be 12 meters or more. However, while these numbers sound impressive, it's worth pointing out that modern tunnel-boring machines can excavate tunnels as wide as 16 meters at rates of up to 4.8 meters per hour. The nuclear subterrene's performance could potentially be substantially improved with better materials and a higher-temperature reactor, but this was not to be.

The Later Years of the Subterrene
In 1975, the program's funding was transferred from the National Science Foundation to the newly born Energy Research and Development Administration. By this point, the funding agencies had directed the project to shift away from giant nuclear subterrenes and towards development of small, electrically-powered penetrators for use in geothermal drilling. In fact, the nuclear subterrene wasn't even mentioned in the program's final status report.

But, without the nuclear reactor, the subterrene simply wasn't economical. LASL looked for partners in industry to commercialize the system, but couldn't find any. The penetrator used up too much energy, it cost too much, and its advantages over existing methods – lower environmental impact, arbitrary tunnel shape – were too slight to justify using it. The program was shut down entirely in 1976.

However, like all technology development programs, the subterrene left heirs – in fact, its first progeny, the Hot Dry Rock program, was born almost immediately after the subterrene's own genesis in 1970. HDR was developed by LASL to explore one of the original applications of the subterrene, producing geothermal energy by drilling into deep, hot rock and using it to heat water. Research at LASL on HDR continued off and on until 1996, and included the construction of a test plant at Fenton Hill on the Los Alamos laboratory grounds.

The subterrene itself was less lucky. Subterrenes continued to occasionally appear in lists of promising new drilling technologies through the early 80s, and various rock-melting drilling techniques continue to be patented and discussed in engineering journals even today. But these are exclusively electrical systems.

However, while the nuclear subterrene was abandoned by the engineering community, it found a home in more unusual places. One was among the theorists of space exploration. In 1986, Dr. John Rowley and two other scientists from the subterrene program published a paper suggesting using nuclear subterrenes to excavate tunnels on the moon to shield colonists from radiation – a subselene, as they called it. Subselenes have continued to intermittently appear in speculative scientific work on lunar colonization.

But they weren't the only people interested in the subterrene. If you google “subterrene,” most of the hits you will find will be websites claiming that the US government completed the subterrene technology, and has used it to excavate a massive complex of hundreds of underground bases linked by tunnels covering most of the American southwest and beyond. Some of the claims made are prodigious – that thousands of deep underground bases have been built across the world, and that photos of what is clearly a tunnel-boring machine used to prepare underground nuclear weapons tests actually depict a nuclear subterrene. This belief appears to stem from a 1995 book, Underground Bases and Tunnels: What is the Government Trying to Hide?, by Richard Sauder. It is an interesting that, not only does most of the “research” appear to be copy & pastes of excerpts from the 1975 subterrene patents and Sauder's book, but they all appear to be the same excerpts, copied from website to website like a chain letter.

But the most creative descendant of the subterrene can only be the proposal to weaponize the system as a way to attack hardened underground military bases, as an alternative to nuking them. The RadioIsotope Powered Thermal Penetrator (RIPTP) does not use a nuclear reactor for heat; instead, it uses Thulium-170 or Ytterbium-168, highly radioactive artificial isotopes, which generate heat as they decay. The RIPTP would be parachuted onto the ground above the base, and would then melt its way through the ground. The motive power to push the subterrene through the ground would be provided by gravity; the penetrator would be about four times as dense as the molten rock in front of it and so would sink through it.

As it bores its way down, the RIPTP would form a bubble of magma and hot, high-pressure gasses behind itself. When it nears the underground base, the pressure of the gas and magma would burst the base walls explosively, destroying facilities near the breach through blast and fire. If the magma wasn't enough, incendiary explosives such as aluminum powder could be added. Then, its work done, the RIPTP would keep borrowing downwards past the target, ultimately entombing itself several hundred meters further down.

But, for all the speculation of secret bases, lunar colonies, and weaponization, a nuclear subterrene has not been built and is not likely to be built. The supposed improvements are simply too small and too uncertain to justify the technological and political risk of such a project. Thus, the subterrene is likely to remain what it is – an interesting but unbuilt idea.

End Note – The Russian Competition
According to wikipedia, the Russians had their own version of the subterrene that they called a “battle mole,” and which they actually tested in the 60s. It's difficult to be sure, since I don't speak Russian and I'm using Google translate, but it looks like there are two pieces to this. The first piece is that there was a real Russian program based on a German idea in the 50s to design an enlarged, crewed tunnel-boring machine, called an “underground boat,” powered by chemical fuel. The second is a claim that, under Khrushchev, the Soviets actually built one of these things, powered by a nuclear reactor, which failed spectacularly during testing.

I strongly suspect that the supposed nuclear-powered prototype was an April Fools' hoax by the Russian-language Popular Mechanics magazine. The first mention of it I can find online comes from the April issue of that magazine, and the diagram they include incorporates what appear to be mechanical tentacles. In any case, however, neither of these were a subterrene, but rather tunnel boring machines.


Atom, The. “The First Practical Application of the Subterrene.” Vol. 10, No. 4, May 1973. Pp. 1 – 4.
Atom, The. “Burning Through-the-Earth Demonstration Intrigues Washington, D. C. Audience.” Vol. 10, No. 9, November-December 1973. Pp. 10 – 12.
Branscome, Ewell Caleb. A Multidisciplinary Approach to the Identification and Evaluation of Novel Concepts for Deeply Buried Hardened Target Defeat. Graduate Thesis, School of Aerospace Engineering, Georgia Institute of Technology, 2006.
Neudecker, Joseph W., James D. Blacic, and John C. Rowley. “Subselene: A Nuclear Powered Melt Tunneling Concept for High-Speed Lunar Subsurface Transportation Tunnels.” Submitted to Symposium '86. Los Alamos National Laboratory, 1986. LA-UR-86-2897.
Rapid Excavation by Rock Melting – LASL Subterrene Program, December 31, 1972, to September 1, 1973.” Comp. Hanold, R. J. Los Alamos Scientific Laboratory, November 1973. LA-5459-SR.
Rapid Excavation by Rock Melting – LASL Subterrene Program, September 1973 – June 1976.” Comp. Hanold, R. J. Los Alamos Scientific Laboratory, 1977. LA-5979-SR.
Robinson, E. S., Rowley, J. C., Potter, R. M, et al. A Preliminary Study of the Nuclear Subterrene. Los Alamos Scientific Laboratory. LA-4547.
Smith, Morton C. The Furnace in the Basement, Part I: The Early Days of the Hot Dry Rock Geothermal Energy Program, 1970 – 1973. Los Alamos National Laboratory, 1995. LA-12809, Part I.
US Patent No. 3,881,777. “Apparatus and Method for Large Tunnel Excavation in Soft and Incompetent Rock or Ground.” Inv. John H. Altseimer and Robert J. Hanold, assigned to The United States of America as represented by the US Energy Research and Development Administration. Filed Jan. 25 1974, pat. May 6 1975.
US Patent No. 3,885,832. “Apparatus and Method for Large Tunnel Excavation in Hard Rock.” Inv. John H. Altseimer and Robert J. Hanold, assigned to the United States of America as represented by the US Energy Research and Development Administration. Filed Jan. 25 1974, pat. May 27 1975.

Imagery Sources:

Figure 1: US Government.   Found in The Furnace in the Basement.
Figure 2: US Government.   Found in "First Practical Application."
Figure 3: US Government.   Found in "From LASL to Industry With Love", The Atom, Vol. 12 No. 6, November/December 1975, p. 1.
Figure 4: US Government.
Figure 5: US Government.   Found in Patent No. 3,881,777.
Figure 6: US Government.   Found in Patent No. 3,885,832.


  1. For a similar, less-developed idea, see the geophysical explorer proposed by William Mansfield Adams in 1961:

    The tantalizing ocean beneath Europa's surface beckons the imaginative subterrene-builder.

    --Bill Higgins

    1. There was also a German gentleman who patented an atomic mining machine of a similar design in the 50s, although I didn't find any evidence anyone else paid any attention to the idea at the time.

      Thanks for reading!

  2. So far this is the only factual account I can find on the "Subterrene". Thanks for the careful research.

  3. VERY nicely done! Attention-holding style and superb research. I lean toward someone did develop the nuclear variety because it just seems like something we'd want. Reading this, I must still search further, but, what a good article! Mind if I quote you almost verbatim and drop a link to your site and this?
    Thanks much...appreciated.

    1. Sure, I don't mind if people quote - even extensively - as long as you link back to me.

      I really doubt that we actually did develop the nuclear system, for two reasons. First, the only potential advantage of the nuclear subterrene over conventional TBMs is cost, and black budget projects are noteworthy in their lack of attention to cost. And second, it's not at all clear that it *would* be any cheaper - this is something I didn't get into in the article, because when I was writing it I didn't know about it, but this research started during the Great Bandwagon Market for nuclear energy. In the mid-late 1970s, the price of nuclear energy spiked sharply and never came back down. So if you actually built this machine, it's not at all clear it would be at all superior to a TBM.

    2. That makes sense, and thanks for the added info.
      What makes me wonder is that the nuclear version could run on and on for a considerable depth, and length of time, as opposed to electrical or ???
      I really don't get the fascination the U.S., China, Russia and others have with drilling to the earth's mantle, but seems like, again, nuclear drills would fare much better than any others.
      CNN did a piece on that in 2012
      Are there enough drill pipes, bits? Would they hold up?
      Remembering the difficulties Western Geophysical oil exploration crews had with under 1000 feet in the desert of Iran, 1966/67. Again, thanks much. I will assuredly link to you.

    3. Although theoretically a nuclear subterrene could operate independently of the surface, in practice it couldn't. Even if you add on crew sleeping compartments and stores of spare parts, and just dump waste rock in the space behind you rather then digging a tunnel, you still need some way to dispose of waste heat. Rock is a very good insulator. If you try to go deep without a surface connection, you'll keep heating up until eventually you melt. This is generally considered suboptimal.

      The goal is not to reach the actual core. That's physically impossible, and will be for the imaginable future. We could probably eventually find something that could survive the temperature and corrosion, but not the pressure - you'd be crushed long before you'd melt. The goal is "only" to sample the upper mantle, which is challenging enough by itself. Even there, the real challenge isn't the drilling technology, it's materials that can hold up to the titanic stresses down there. So the subterrene wouldn't help much in reaching the mantle.

      The reason we want to reach the Earth's mantle is because the chemical and isotopic composition of the mantle can tell us a lot about the Earth's geological formation and evolution. We can get samples from volcanoes, but the magma in those has probably changed in unknowable ways on its journey through the crust. For the scientists, it's knowledge for its' own sake. For the funding agencies, it's a combination of that, and the fact that better understanding of geology will probably ultimately lead to applications in mining, geothermal energy, and other areas.