Going Interstellar (22 page)

Read Going Interstellar Online

Authors: Les Johnson,Jack McDevitt

BOOK: Going Interstellar
6.31Mb size Format: txt, pdf, ePub

Before the end of the Cold War (during which thousands of fission and fusion devices were produced) futurists realized that human civilization would ultimately exhaust its fossil-fuel reserves. Perhaps some form of controlled thermonuclear fusion might be the answer to our growing energy needs.

Two basic types of electricity-producing fusion reactors have been proposed and are being researched. One approach uses powerful electric and magnetic fields to confine the plasma (ionized gas) of thermonuclear material. Another major difficulty in achieving controlled thermonuclear fusion is the multi-million-degree temperature at which the reactants must be maintained. Although cleaner (in terms of radioactivity) than the less-powerful fission reactors now in use, currently feasible fusion reactors will also produce some radioactivity.

Confined-fusion technologists use two benchmarks to define their progress. Achievement of “scientific breakeven” would mean that an experimental fusion reactor would produce as much output energy as was used to create the fusion reaction to begin with. “Technological breakeven” means that the energy produced is at least ten times greater than the energy input. At present, experimental confined-fusion reactors operate at about 50% of scientific breakeven. Achievement of technological breakeven will require more time—and money.

Although confined-fusion reactors have promise for terrestrial energy production, inertial fusion might be more useful for in-space propulsion. Inertial fusion reactors operate using small pellets of fusion reactants. These are pelted with electron beams or lasers to raise pellet temperature and density to levels at which thermonuclear reactions can occur. Essentially, an inertial-fusion reactor is a small hydrogen bomb with the fission trigger replaced by electron or laser beams.

An inertial-fusion reactor used to produce terrestrial energy would require considerable shielding to trap the high-energy products of the thermonuclear reactions. But this is less of a problem in space. Since these reaction products largely consist of high-energy electrically charged particles, engineers quickly figured out that they could simply squirt them out the back of the spacecraft as rocket exhaust. Even before Apollo 11 reached the Moon, some scientists realized that inertial-fusion ships might some day reach the stars!

 

 

Project Orion—Birth of the Interstellar Dream

 

Freeman Dyson distrusted bureaucracies. During the Second World War, he worked on crew safety for the British Royal Air Force Bomber Command. Early in the war, he realized that the escape hatches on many British bombers were too small for crewmembers to depart a stricken aircraft while wearing their parachutes. Dyson wrote memo after memo to correct this defect without positive response until late in the war. Embittered, he realized that thousands of brave British airmen must have needlessly perished. He swore that never again would he trust a large bureaucracy to do the right thing. More than anything else, Dyson’s response to his wartime experience helped produce the realization that the stars are not beyond reach.

After the war, when Dyson had moved to the Princeton University Institute of Advanced Study, he mentored Theodore Taylor in his Ph.D. studies. Working on the US atomic bomb project, Taylor had become disillusioned with the effort that went into creating fake cities and nuking them. To him, this was a waste of taxpayer money since the A-bomb, after all, had been “tested” on two very real Japanese cities. Taylor, instead of concentrating on the construction of objects to be destroyed by atomic blasts, asked himself if anything could survive in the hellish vicinity near ground zero.

He designed a pumpkin-sized steel sphere, coated it with graphite, and installed it at the Eniwetok nuclear test site in the Pacific near a 20-kiloton nuclear device. To everyone’s surprise (but perhaps not Taylor’s), the metal sphere rode out the blast with minimal damage. Apparently, the graphite layer had ablated — evaporating at high speed — and carried off much of the incident energy produced by the explosion.

Dyson, Taylor and others saw a possible application for this process. As the Space Age dawned, US defense analysts recognized that there was no known defense against orbital Soviet nuclear warheads. But perhaps a spacecraft propelled by external nuclear explosions might do the trick.

This was the birth of the initially top-secret Project Orion. On a future spacecraft, Orion crews would carry with them small nuclear charges. (Okay, they would be small bombs.) The charges would be discharged on command behind a pusher plate coated with ablative material. This pusher plate, which would be impacted by the nuclear blast, would be connected to the rest of the ship by the world’s largest shock absorbers.
Bang! Bang! Bang!
Explosion after explosion would impulsively propel spacecraft to faster and faster speeds.

Although a full-scale Orion was never constructed, small test models propelled by chemical explosives were successfully filmed careening across the sky. One is on display (near a model of Star Trek’s Starship
Enterprise
) in the Smithsonian Air and Space Museum in Washington D.C.

As the Project Orion study continued, it became evident that Orion “interceptors” could be capable of velocities in excess of 30 miles (50 kilometers) per second. Some conceptual versions could lift from Earth under their own nuclear drive, unfortunately leaving behind a huge wake of radioactive particles. Variants might ride as the second stage of a Saturn V rocket, exhausting their A-bombs well above Earth’s delicate biosphere.

The high time for Project Orion was in 1961-1963. NASA had been commissioned by President Kennedy to deliver and return humans from the Moon before 1970. Most analysts preferred the Saturn V booster to launch the Moon ships, but this rocket had not yet been tested. So a number of back-ups were suggested. One was Orion.

In this heady period, Dyson, Taylor and their associates investigated the interplanetary potential of Orion. As a Saturn V upper stage, it had the potential of ferrying astronauts to Mars on month-long journeys. Habitats, rovers, greenhouses and livestock could come along as well.

But alas, it was not to be. The Atmospheric Test Ban Treaty dampened the prospects for Orion. And the success of Saturn V doomed it. Before the first Lunar Modules swooped down over the lunar plains, Orion and its extensive documentation seemed headed for storage in some super-secret government depository, perhaps located next to the box containing Indiana Jones’s Ark of the Covenant.

Freeman Dyson was angry. And Freeman Dyson distrusted large government bureaucracies. So he methodically hatched a scheme to save Project Orion from oblivion.

Being a physicist, Dyson planned to publish a paper describing the potential of Orion in a journal. But most physics, astronomy, and astronautics journals have circulations of only a few thousand. He chose to publish in
Physics Today
, a semi-popular monthly organ of The American Institute of Physics. Many public and university libraries subscribe to this magazine—its monthly readership would therefore be much larger than that of more technical physics journals. Dyson planned a paper that would outline the concept of Orion in visionary terms, and do so in a manner that would not violate his oath of secrecy.

Of course he had to use clever approximations. One was the yield in equivalent megatons of TNT of a deuterium-fueled thermonuclear explosive. Dyson knew that the USSR had just air tested the largest H-bomb ever exploded. The yield of the test was well established and the type of aircraft carrying the device had been announced. Dyson probably could have exactly stated the yield of a fusion explosive—instead, he consulted a standard reference (
Jane’s All the World’s Aircraft
) and used the payload capacity of the Soviet bomber.

Published in late 1968, Dyson’s paper established him as an early hero of the “Interstellar Movement.” Even with his many approximations, he demonstrated that huge, multi-kilometer fusion-pulse world ships could be constructed that would take up to one thousand years to reach the nearest stars. If the entire US/USSR 1968-vintage thermonuclear arsenals had been devoted to Project Orion, as many as 20,000 people could have been relocated to the Alpha / Proxima Centauri system. What a happy use for the bombs!

 

 

Projects
Daedalus
and
Icarus
—The BIS follows Up

 

Now that Dyson and Taylor had opened the “Interstellar Door,” other groups began their own studies. The British Interplanetary Society (BIS), which had studied Moon flight decades before the Apollo Project, was ideally situated to conduct a follow-on study to Orion. British researchers Alan Bond and Anthony Martin directed this study, dubbed Project
Daedalus
, during the 1970’s. The original
Daedalus
, a mythological Athenian architect, had escaped imprisonment in Crete with the aid of flapping wings handily crafted from goose feathers.

It was soon determined that the modern
Daedalus
, although inspired by the Orion conceptual breakthrough, would be a bit different. Several problems were acknowledged with the Orion concept. One was scale—an Orion starship (such as the pulsed thermonuclear rocket shown schematically in Figure 2) would be huge even if its payload were small. This was due to the size of the equipment necessary to deflect the copious particles emitted by even a small thermonuclear blast. Another issue was psychological—how would the crew and passengers of a starship react to a megaton-sized explosion going off every few seconds, at a distance of only a kilometer or so? Finally, it is difficult to conceive of any real-world scenario in which nuclear superpowers would allow use of their arsenals in such a constructive endeavor.

 

 

Figure 2. Artist concept of a Project Orion nuclear pulse spacecraft. (Image courtesy of NASA.)

 

Daedalus
evolved as a kid brother to Orion. Instead of using the dramatic thermonuclear-pulse drive, it used a somewhat tamer approach—inertial fusion. Small micropellets of fusion fuel were to be ejected into a combustion chamber equipped with strong magnetic fields. Instead of ignition by a fission trigger, these pellets were to be heated to fusion temperatures and condensed to fusion densities by an array of focused laser or electron beams.

Researchers involved in the effort spent a good deal of time considering fusion fuel cycles. They rejected the deuterium-tritium (D-T) and deuterium-deuterium (D-D) fusion reactions under active consideration for terrestrial energy production. Although cleaner than fission, the copious thermal neutrons produced by these reactions would rapidly irradiate the spacecraft. Instead, they settled on a reaction between a low-mass form of helium (Helium-3) and deuterium. The products of this reaction are electrically charged particles—these are relatively easy to focus and expel with the aid of powerful magnetic fields.

Although the Helium-3/D reaction is the second easiest to ignite after D-T, it has one significant drawback. Helium-3 is very, very rare in the terrestrial environment. Starship designers were faced with four alternatives to obtain the necessary tens of millions of kilograms of this substance.

 

1. They could pepper the surface of the Earth or Moon with breeder reactors, which produce more nuclear fuel than they consume to produce it.
2. Since Helium-3 is a trace component of the solar wind of ions ejected from our Sun, some form of superconducting electromagnetic scoop could mine the solar wind for this isotope—but high temperatures in the inner solar system might render superconducting scoops difficult to build and maintain.
3. Tiny amounts of He-3 had been deposited in the upper layers of lunar soil as evidenced by samples returned by Apollo astronauts—but at that time nobody knew how the He-3 concentration varied with depth in lunar soils and how feasible lunar mining might actually be.
4. What they opted for was the fourth alternative: He-3 is found in the atmospheres of giant planets. Perhaps a series of robotic helium mines suspended by balloons in the upper atmosphere of Jupiter would be the answer.

 

Although the
Daedalus
engine could in concept be used to accelerate and decelerate a “thousand-year ark,” the initial application was expected to be robotic probes that could be accelerated to about 10% the speed of light (0.1c) and then fly through the destination star system. In the 1970s, it was (erroneously) suspected that the second nearest star—a red dwarf called Barnard’s star at a distance of about six light years from the Sun—had Jupiter-sized planets. So Barnard’s Star was selected for the hypothetical star mission.

Project
Daedalus
resulted in and inspired many papers published in dedicated issues of
JBIS
(
The Journal of the British Interplanetary Society
). In 2010, a follow-up BIS study called Project
Icarus
(after the son of mythological Daedalus who approached the Sun too closely and fell to his death in the Aegean) commenced.

Other books

The Sharecropper Prodigy by Malone, David Lee
New World, New Love by Rosalind Laker
Murder Fortissimo by Nicola Slade
Bombora by Mal Peters
Rocked by Bayard, Clara
Taken by Barbara Freethy
Gambler by S.J. Bryant