Physics of the Impossible

Chapter 7

Obviously a new kind of rocket design must be found if we are ever to reach the stars. Either we must radically increase the thrust of our rockets, or we need to increase the time over which our rockets operate. A large chemical rocket, for example, may have the thrust of several million pounds, but it burns for only a few minutes. By contrast, other rocket designs, such as the ion engine (described in the following paragraphs), may have a feeble thrust but can operate for years in outer s.p.a.ce. When it comes to rocketry, the tortoise wins over the hare.

ION AND PLASMA ENGINES.

Unlike chemical rockets, ion engines do not produce the sudden, dramatic blast of superhot gases that propel conventional rockets. In fact, their thrust is often measured in ounces. Placed on a tabletop on Earth, they are too feeble to move. But what they lack in thrust they more than make up for in duration, because they can operate for years in the vacuum of outer s.p.a.ce.

A typical ion engine looks like the inside of a TV tube. A hot filament is heated by an electric current, which creates a beam of ionized atoms, such as xenon, that is shot out the end of the rocket. Instead of riding on a blast of hot, explosive gas, ion engines ride on a thin but steady flow of ions.

NASA"s NSTAR ion thruster was tested in outer s.p.a.ce aboard the successful Deep s.p.a.ce 1 probe, launched in 1998. The ion engine fired for a total of 678 days, setting a new record for ion engines. The European s.p.a.ce Agency has also tested an ion engine on its Smart 1 probe. The j.a.panese Hayabusa s.p.a.ce probe, which flew past an asteroid, was powered by four xenon ion engines. Although unglamorous, the ion engine will be able to make long-haul missions (that are not urgent) between the planets. In fact, ion engines may one day become the workhorse for interplanetary transport.

A more powerful version of the ion engine is the plasma engine, for example, the VASIMR (variable specific impulse magnetoplasma rocket), which uses a powerful jet of plasma to propel it through s.p.a.ce. Designed by astronaut/engineer Franklin Chang-Diaz, it uses radio waves and magnetic fields to heat hydrogen gas to a million degrees centigrade. The superhot plasma is then ejected out the end of the rocket, yielding significant thrust. Prototypes of the engine have already been built on Earth, although none has ever been sent into s.p.a.ce. Some engineers hope the plasma engine can be used to power a mission to Mars, significantly reducing the travel time to Mars, down to a few months. Some designs use solar power to energize the plasma in the engine. Other designs use nuclear fission (which raises safety concerns, since it involves putting large amounts of nuclear materials into s.p.a.ce on ships that are susceptible to accident).

Neither the ion nor the plasma/VASIMR engine, however, has enough power to take us to the stars. For that, we need an entirely new set of propulsion designs. One serious drawback to designing a starship is the staggering amount of fuel necessary to make a trip to even the nearest star, and the long span of time before the ship reaches its distant destination.

SOLAR SAILS.

One proposal that may solve these problems is the solar sail. It exploits the fact that sunlight exerts a very small but steady pressure that is sufficient to propel a huge sail through s.p.a.ce. The idea for a solar sail is an old one, dating back to the great astronomer Johannes Kepler in his 1611 treatise Somnium.

Although the physics behind a solar sail is simple enough, progress has been spotty in actually creating a solar sail that can be sent into s.p.a.ce. In 2004 a j.a.panese rocket successfully deployed two small prototype solar sails into s.p.a.ce. In 2005 the Planetary Society, Cosmos Studios, and the Russian Academy of Sciences launched the Cosmos 1 s.p.a.ce sail from a submarine in the Barents Sea, but the Volna rocket it was being carried on failed, and the sail did not reach orbit. (A previous attempt at a suborbital sail also failed back in 2001.) But in February 2006 a 15-meter solar sail was sent successfully into orbit by the j.a.panese M-V rocket, although the sail opened incompletely.

Although progress in solar sail technology has been painfully slow, proponents of the solar sail have another idea that might take them to the stars: building a huge battery of lasers on the moon that can fire intense beams of laser light at a solar sail, enabling it to coast to the nearest star. The physics of such an interplanetary solar sail are truly daunting. The sail itself would have to be hundreds of miles across and constructed entirely in outer s.p.a.ce. One would have to build thousands of powerful laser beams on the moon, each capable of firing continuously for years to decades. (In one estimate, it would be necessary to fire lasers that have one thousand times the current total power output of the planet Earth.) On paper a mammoth light sail might be able to travel as fast as half the speed of light. It would take such a solar sail only eight years or so to reach the nearby stars. The advantage of such a propulsion system is that it could use off-the-shelf technology. No new laws of physics would have to be discovered to create such a solar sail. But the main problems are economic and technical. The engineering problems in creating a sail hundreds of miles across, energized by thousands of powerful laser beams placed on the moon, are formidable, requiring a technology that may be a century in the future. (One problem with the interstellar solar sail is coming back. One would have to create a second battery of laser beams on a distant moon to propel the vessel back to Earth. Or perhaps the ship could swing rapidly around a star, using it like a slingshot to get enough speed for the return voyage. Then lasers on the moon would be used to decelerate the sail so it could land on the Earth.) RAMJET FUSION.

My own favorite candidate for getting us to the stars is the ramjet fusion engine. There is an abundance of hydrogen in the universe, so a ramjet engine could scoop hydrogen as it traveled in outer s.p.a.ce, essentially giving it an inexhaustible source of rocket fuel. Once the hydrogen was collected it would then be heated to millions of degrees, hot enough so that the hydrogen would fuse, releasing the energy of a thermonuclear reaction.

The ramjet fusion engine was proposed by physicist Robert W. Bussard in 1960 and later popularized by Carl Sagan. Bussard calculated that a ramjet engine weighing about 1,000 tons might theoretically be able to maintain a steady thrust of 1 g of force, that is, comparable to standing on the surface of the Earth. If the ramjet engine could maintain a 1 g acceleration for one year, it would reach 77 percent of the velocity of light, sufficient to make interstellar travel a serious possibility.

The requirements for the ramjet fusion engine are easy to compute. First, we know the average density of hydrogen gas throughout the universe. We also can calculate roughly how much hydrogen gas must be burned in order to attain 1 g accelerations. That calculation, in turn, determines how big the "scoop" must be in order to gather hydrogen gas. With a few reasonable a.s.sumptions, one can show that you would need a scoop that is about 160 kilometers in diameter. Although creating a scoop of this size would be prohibitive on Earth, building it in outer s.p.a.ce poses fewer problems because of weightlessness.

In principle the ramjet engine could propel itself indefinitely, ultimately reaching distant star systems in the galaxy. Since time slows down inside the rocket, according to Einstein, it might be possible to reach astronomical distances without resorting to putting the crew into suspended animation. After accelerating at 1 g for eleven years, according to clocks inside the starship, the s.p.a.cecraft would reach the Pleiades star cl.u.s.ter, which is 400 light-years away. In twenty-three years it would reach the Andromeda galaxy, which is 2 million light-years from Earth. In theory, the s.p.a.cecraft might be able to reach the limit of the visible universe within the lifetime of a crew member (although billions of years might have pa.s.sed on the Earth).

One key uncertainty is the fusion reaction. The ITER fusion reactor, scheduled to be built in the south of France, combines two rare forms of hydrogen (deuterium and tritium) in order to extract energy. In outer s.p.a.ce, however, the most abundant form of hydrogen consists of a single proton surrounded by an electron. The ramjet fusion engine would therefore have to exploit the proton-proton fusion reaction. Although the deuterium/tritium fusion process has been studied for decades by physicists, the proton-proton fusion process is less well understood, is more difficult to achieve, and yields far less power. So mastering the more difficult proton-proton reaction will be a technical challenge in the coming decades. (Some engineers, in addition, have questioned whether the ramjet engine could overcome drag effects as it approaches the speed of light.) Until the physics and economics of proton-proton fusion are worked out, it is difficult to make accurate estimates as to the ramjet"s feasibility. But this design is on the short list of possible candidates for any mission contemplated to the stars.

NUCLEAR ELECTRIC ROCKET.

In 1956 the U.S. Atomic Energy Commission (AEC) began to look at nuclear rockets seriously under Project Rover. In theory, a nuclear fission reactor would be used to heat up gases like hydrogen to extreme temperatures, and then these gases would be ejected out one end of the rocket, creating thrust.

Because of the danger of an explosion in the Earth"s atmosphere involving toxic nuclear fuel, early versions of nuclear rocket engines were placed horizontally on railroad tracks, where the performance of the rocket could be carefully monitored. The first nuclear rocket engine to be tested under Project Rover was the Kiwi 1 in 1959 (aptly named after the Australian flightless bird). In the 1960s NASA joined with the AEC to create the Nuclear Engine for Rocket Vehicle Applications (NERVA), which was the first nuclear rocket to be tested vertically, rather than horizontally. In 1968 this nuclear rocket was test-fired in a downward position.

The results of this research have been mixed. The rockets were very complicated and often misfired. The intense vibrations of the nuclear engine often cracked the fuel bundles, causing the ship to break apart. Corrosion due to burning hydrogen at high temperatures was also a persistent problem. The nuclear rocket program was finally closed in 1972.

(These atomic rockets had yet another problem: the danger of a runaway nuclear reaction, as in a small atomic bomb. Although commercial nuclear power plants today run on diluted nuclear fuel and cannot explode like a Hiroshima bomb, these atomic rockets, in order to create maximum thrust, operated on highly enriched uranium and hence could explode in a chain reaction, creating a tiny nuclear detonation. When the nuclear rocket program was about to be retired, scientists decided to perform one last test. They decided to blow up a rocket, like a small atomic bomb. They removed the control rods [which keep the nuclear reaction in check]. The reactor went super-critical and blew up in a fiery ball of flames. This spectacular demise of the nuclear rocket program was even captured on film. The Russians were not pleased. They considered this stunt to be a violation of the Limited Test Ban Treaty, which banned above-ground detonations of nuclear bombs.) Over the years the military has periodically revisited the nuclear rocket. One secret project was called the Timberwind nuclear rocket; it was part of the military"s Star Wars project in the 1980s. (It was abandoned after details of its existence were released by the Federation of American Scientists.) The main concern about the nuclear fission rocket is safety. Even fifty years into the s.p.a.ce age, chemical booster rockets undergo catastrophic failure about 1 percent of the time. (The two failures of the Challenger and Columbia s.p.a.ce Shuttles, tragically killing fourteen astronauts, further confirmed this failure rate.) Nonetheless, in the past few years NASA has resumed research on the nuclear rocket for the first time since the NERVA program of the 1960s. In 2003 NASA christened a new project, Prometheus, named for the Greek G.o.d who gave fire to humanity. In 2005 Prometheus was funded at $430 million, although that funding was significantly cut to $100 million in 2006. The project"s future is unclear.

NUCLEAR PULSED ROCKETS.

Another distant possibility is to use a series of mini-nuclear bombs to propel a starship. In Project Orion, mini-atomic bombs were to be ejected out the back of the rocket in sequence, so that the s.p.a.cecraft would "ride" on the shock waves created by these mini-hydrogen bombs. On paper such a design could take a s.p.a.cecraft close to the speed of light. Originally conceived in 1947 by Stanislaw Ulam, who helped to design the first hydrogen bombs, the idea was further developed by Ted Taylor (one of the chief designers of nuclear warheads for the U.S. military) and physicist Freeman Dyson of the Inst.i.tute for Advanced Study at Princeton.

In the late 1950s and 1960s elaborate calculations were made for this interstellar rocket. It was estimated that such a starship could make it to Pluto and back within a year, with a top cruising velocity of 10 percent the speed of light. But even at that speed it would take about forty-four years to reach the nearest star. Scientists have speculated that a s.p.a.ce ark powered by such a rocket would have to cruise for centuries, with a multigenerational crew whose offspring would be born and spend all their lives on the s.p.a.ce ark, in order that their descendants could reach the nearby stars.

In 1959 General Atomics issued a report estimating the size of an Orion s.p.a.cecraft. The largest version, called the super Orion, would weigh 8 million tons, have a diameter of 400 meters, and be energized by over 1,000 hydrogen bombs.

But one major problem with the project was the possibility of contamination via nuclear fallout during launch. Dyson estimated that the nuclear fallout from each launch could cause fatal cancers in ten people. In addition, the electromagnetic pulse (EMP) for such a launch would be so great that it could cause ma.s.sive short circuits in neighboring electrical systems.

The signing of the Limited Test Ban Treaty in 1963 sounded the death knell of the project. Eventually the main driving force pushing the project, nuclear bomb designer Ted Taylor, gave up. (He once confided to me that he finally became disillusioned with the project when he realized that the physics behind mini-nuclear bombs could also be used by terrorists to create portable nuclear bombs. Although the project was canceled because it was deemed too dangerous, its namesake lives on in the Orion s.p.a.cecraft, which NASA has chosen to replace the s.p.a.ce Shuttle in 2010.) The concept of a nuclear-fired rocket was briefly resurrected by the British Interplanetary Society from 1973 to 1978, with Project Daedalus, a preliminary study to see if an unmanned starship could be built that could reach the Barnard"s Star, 5.9 light-years from Earth. (Barnard"s Star was chosen because it was conjectured that it might have a planet. Since then astronomers Jill Tarter and Margaret Turnbull have compiled a list of 17,129 nearby stars that could have planets supporting life. The most promising candidate is Epsilon Indi A, 11.8 light-years away.) The rocket ship planned for Project Daedalus was so huge that it would have had to be constructed in outer s.p.a.ce. It would weigh 54,000 tons, nearly all of its weight in rocket fuel, and could attain 7.1 percent of the speed of light with a payload of 450 tons. Unlike Project Orion, which used tiny fission bombs, Project Daedalus would use mini-hydrogen bombs with a deuterium/helium-3 mixture ignited by electron beams. Because of the formidable technical problems facing it, as well as concerns over its nuclear propulsion system, Project Daedalus was also shelved indefinitely.

SPECIFIC IMPULSE AND ENGINE EFFICIENCY.

Engineers sometimes speak of "specific impulse," which enables us to rank the efficiency of various engine designs. "Specific impulse" is defined as the change in momentum per unit ma.s.s of propellant. Hence the more efficient the engine, the less fuel is necessary to boost a rocket into s.p.a.ce. Momentum, in turn, is the product of the force acting over a period of time. Chemical rockets, although they have very large thrust, operate for only a few minutes, and hence have a very low specific impulse. Ion engines, because they can operate for years, can have high specific impulse with very low thrust.

Specific impulse is measured in seconds. A typical chemical rocket might have a specific impulse of 400500 seconds. The specific impulse of the s.p.a.ce Shuttle engine is 453 seconds. (The highest specific impulse ever achieved for a chemical rocket was 542 seconds, using a propellant mixture of hydrogen, lithium, and fluorine.) The thruster for the Smart 1 ion engine had a specific impulse of 1,640 seconds. And the nuclear rocket attained specific impulses of 850 seconds.

The maximum possible specific impulse would be a rocket that could attain the speed of light. It would have a specific impulse of about 30 million. Following is a table showing the specific impulses of different kinds of rocket engines.

TYPE OF ROCKET ENGINE.

SPECIFIC IMPULSE.

Solid fuel rocket 250.

Liquid fuel rocket 450.

Ion engine 3,000.

VASIMR plasma engine 1,000 to 30,000 Nuclear fission rocket 800 to 1,000 Nuclear fusion rocket 2,500 to 200,000 Nuclear pulsed rocket 10,000 to 1 million Antimatter rocket.

1 million to 10 million.

(In principle, laser sails and ram-jet engines, because they contain no rocket propellant at all, have infinite specific impulse, although they have problems of their own.) s.p.a.cE ELEVATORS.

One severe objection to many of these rocket designs is that they are so mammoth and heavy that they could never be built on the Earth. That is why some scientists have proposed building them in outer s.p.a.ce, where weightlessness would make it possible for astronauts to lift impossibly heavy objects with ease. But critics today point out the prohibitive costs of a.s.sembly in outer s.p.a.ce. The International s.p.a.ce Station, for example, will require upwards of one hundred launches of shuttle missions for complete a.s.sembly and costs have escalated to $100 billion. It is the most expensive scientific project in history. Building an interstellar s.p.a.ce sail or ramjet scoop in outer s.p.a.ce would cost many times that amount.

But as science fiction writer Robert Heinlein was fond of saying, if you can make it to 160 kilometers above the Earth, you are halfway to anywhere in the solar system. That is because the first 160 kilometers of any launch, when the rocket is struggling to escape the Earth"s gravity, cost by far the most. After that a rocket ship can almost coast to Pluto and beyond.

One way to reduce costs drastically in the future would be to develop a s.p.a.ce elevator. The idea of climbing a rope to heaven is an old one, for example, as in the fairy tale "Jack and the Beanstalk," but it might become a reality if the rope could be sent far into s.p.a.ce. Then the centrifugal force of the Earth"s rotation would be enough to nullify the force of gravity, so the rope would never fall. The rope would magically rise vertically into the air and disappear into the clouds. (Think of a ball spinning on a string. The ball seems to defy gravity, because the centrifugal force pushes it away from the center of rotation. In the same way, a very long rope would be suspended in air because of the spinning of the Earth.) Nothing would be needed to hold up the rope except the spin of the Earth. A person could theoretically climb the rope and ascend into s.p.a.ce. We sometimes give the undergraduates taking physics courses at City University of New York the problem of calculating the tension on such a rope. It is easy to show that the tension on the rope would be enough to snap even a steel cable, which is why building a s.p.a.ce elevator has long been considered to be impossible.

The first scientist to seriously study the s.p.a.ce elevator was Russian visionary scientist Konstantin Tsiolkovsky. In 1895, inspired by the Eiffel Tower, he envisioned a tower that would ascend into s.p.a.ce, connecting the Earth to a "celestial castle" in s.p.a.ce. It would be built bottom-up, starting on Earth, and engineers would slowly extend the s.p.a.ce elevator to the heavens.

In 1957 Russian scientist Yuri Artsutanov proposed a new solution, that the s.p.a.ce elevator be built in reverse order, top-down, starting from outer s.p.a.ce. He envisioned a satellite in a geostationary orbit 36,000 miles in s.p.a.ce, where it would appear to be stationary, and from which one would drop a cable down to Earth. Then the cable would be anch.o.r.ed to the ground. But the tether for a s.p.a.ce elevator would have to be able to withstand roughly 60100 gigapascals (gpa) of tension. Steel breaks at about 2 gpa, making the idea beyond reach.

The idea of a s.p.a.ce elevator reached a much wider audience with the publication of Arthur C. Clarke"s 1979 novel, The Fountains of Paradise, and Robert Heinlein"s 1982 novel, Friday. But without any further progress, the idea languished.

The equation changed significantly when carbon nanotubes were developed by chemists. Interest was suddenly sparked by the work of Sumio Iijima of Nippon Electric in 1991 (although evidence for carbon nanotubes actually dates back to the 1950s, a fact that was ignored at the time). Remarkably, nanotubes are much stronger than steel cables, but also much lighter. In fact, they exceed the strength necessary to maintain a s.p.a.ce elevator. Scientists believe a carbon nanotube fiber could withstand 120 gpa of pressure, which is comfortably above the breaking point. This discovery has rekindled attempts to create a s.p.a.ce elevator.

In 1999 a NASA study gave serious consideration to the s.p.a.ce elevator, envisioning a ribbon, about 1 meter wide and about 47,000 kilometers long, capable of transporting about 15 tons of payload into Earth"s...o...b..t. Such a s.p.a.ce elevator could change the economics of s.p.a.ce travel overnight. The cost could be reduced by a factor of ten thousand, an astonishing, revolutionary change.

Currently it costs $10,000 or more to send a pound of material into orbit around the Earth (roughly the cost, ounce for ounce, of gold). Each s.p.a.ce Shuttle mission, for example, costs up to $700 million. A s.p.a.ce elevator could reduce the cost to as little as $1 per pound. Such a radical reduction in the cost of the s.p.a.ce program could revolutionize the way we view s.p.a.ce travel. With a simple push of an elevator b.u.t.ton, one could in principle take an elevator ride into outer s.p.a.ce for the price of a plane ticket.

But formidable practical hurdles have to be solved before we build a s.p.a.ce elevator on which we can levitate our way into heaven. At present pure carbon nanotube fibers created in the lab are no more than 15 millimeters long. To create a s.p.a.ce elevator, one would have to create cables of carbon nanotubes that are thousands of miles long. Although from a scientific point of view this is just a technical problem, it is a stubborn and difficult problem that must be solved if we are to create a s.p.a.ce elevator. Yet, within a few decades, many scientists believe that we should be able to master the technology of creating long cables of carbon nanotubes.

Second, microscopic impurities in the carbon nanotubes could make a long cable problematic. Nicola Pugno of the Polytechnic of Turin, Italy, estimates that if a carbon nanotube has even one atom misaligned, its strength could be reduced by 30 percent. Overall, atomic-scale defects could reduce the strength of the nanotube cable by as much as 70 percent, taking it below the minimum gigapascals of strength necessary to support a s.p.a.ce elevator.

To spur entrepreneurial interest in the s.p.a.ce elevator, NASA is funding two separate prizes. (The prizes are modeled on the $10 million Ansari X-prize, which successfully spurred enterprising inventors to create commercial rockets capable of taking pa.s.sengers to the very edge of s.p.a.ce. The X-prize was won by s.p.a.ceship One in 2004.) The prizes NASA is offering are called the Beam Power Challenge and the Tether Challenge. In the Beam Power Challenge, teams have to send a mechanical device weighing at least 25 kilograms up a tether (suspended from a crane) at the speed of 1 meter per second for a distance of 50 meters. This may sound easy, but the catch is that the device cannot use fuel, batteries, or an electrical cord. Instead, the robot device must be powered by solar arrays, solar reflectors, lasers, or microwaves-energy sources that are more suitable for use in outer s.p.a.ce.

In the Tether Challenge, teams must produce 2-meter-long tethers that cannot weigh more than 2 grams and must carry 50 percent more weight than the best tether of the previous year. The challenge is intended to stimulate research in developing lightweight materials strong enough to be strung 100,000 kilometers in s.p.a.ce. There are prizes worth $150,000, $40,000, and $10,000. (To highlight the difficulty of mastering this challenge, in 2005, the first year of the compet.i.tion, no one won a prize.) Although a successful s.p.a.ce elevator could revolutionize the s.p.a.ce program, such machines have their own sets of hazards. For example, the trajectory of near-Earth satellites constantly shifts as they orbit the Earth (this is because the Earth rotates beneath them). This means that these satellites would eventually collide with the s.p.a.ce elevator at 18,000 miles per hour, sufficient to rupture the tether. To prevent such a catastrophe, in the future either satellites will have to be designed to include small rockets so that they can maneuver around the s.p.a.ce elevator, or the tether of the elevator might have to be equipped with small rockets to evade pa.s.sing satellites.

Also, collisions with micrometeorites are a problem, since the s.p.a.ce elevator is far above the atmosphere of the Earth, and our atmosphere usually protects us from meteors. Since micrometeor collisions are unpredictable, the s.p.a.ce elevator must be built with added shielding and perhaps even fail-safe redundancy systems. Problems could also emerge from the effects of turbulent weather patterns on the Earth, such as hurricanes, tidal waves, and storms.

THE SLINGSHOT EFFECT.

Another novel means of hurling an object near the speed of light is to use the "slingshot" effect. When sending s.p.a.ce probes to the outer planets, NASA sometimes whips them around a neighboring planet, so they use the slingshot effect to boost their velocity. NASA saves on valuable rocket fuel in this way. That"s how the Voyager s.p.a.cecraft was able to reach Neptune, which lies near the very edge of the solar system.

Princeton physicist Freeman Dyson proposed that in the far future we might find two neutron stars that are revolving around each other at great speed. By traveling extremely close to one of these neutron stars, we could whip around it and then be hurled into s.p.a.ce at speeds approaching a third the speed of light. In effect, we would be using gravity to give us an additional boost to nearly the speed of light. On paper this just might work.

Others have proposed that we whip around our own sun in order to accelerate to near the speed of light. This method, in fact, was used in Star Trek IV: The Voyage Home, when the crew of the Enterprise hijacked a Klingon ship and then sped close to the Sun in order to break the light barrier and go back in time. In the movie When Worlds Collide, when Earth is threatened by a collision with an asteroid, scientists flee the Earth by creating a gigantic roller coaster. A rocket ship descends the roller coaster, gaining great velocity, and then whips around the bottom of the roller coaster to blast off into s.p.a.ce.

In fact, however, neither of these methods of using gravity to boost our way into s.p.a.ce will work. (Because of the conservation of energy, in going down a roller coaster and coming back up, we wind up with the same velocity as that with which we started, so there is no gain in energy whatsoever. Likewise, by whipping around the stationary sun, we wind up with the same velocity as that with which we originally started.) The reason Dyson"s method of using two neutron stars might work is because the neutron stars are revolving so fast. A s.p.a.cecraft using the slingshot effect gains its energy from the motion of a planet or star. If they are stationary, there is no slingshot effect at all.

Although Dyson"s proposal could work, it does not help today"s Earth-bound scientists, because we would need a starship just to visit rotating neutron stars.

RAIL GUNS TO THE HEAVENS.

Yet another ingenious method for flinging objects into s.p.a.ce at fantastic velocities is the rail gun, which Arthur C. Clarke and others have featured in their science fiction tales, and which is also being seriously examined as part of the Star Wars missile shield.

Instead of using rocket fuel or gunpowder to boost a projectile to high velocity, a rail gun uses the power of electromagnetism.

In its simplest form, a rail gun consists of two parallel wires or rails, with a projectile that straddles both wires, forming a U-shaped configuration. Even Michael Faraday knew that a current of electricity will experience a force when placed in a magnetic field. (This, in fact, is the basis of all electrical motors.) By sending millions of amperes of electrical power down these wires and through the projectile, a huge magnetic field is created around the rails. This magnetic field then propels the projectile down the rails at enormous velocities.

Rail guns have successfully fired metal objects at enormous velocities over extremely short distances. Remarkably, in theory, a simple rail gun should be able to fire a metal projectile at 18,000 miles per hour, so that it would go into orbit around the Earth. In principle, NASA"s entire rocket fleet could be replaced by rail guns that could blast payloads into orbit from the Earth.

The rail gun enjoys a significant advantage over chemical rockets and guns. In a rifle the ultimate velocity at which expanding gases can push a bullet is limited by the speed of shock waves. Although Jules Verne used gunpowder to blast astronauts to the moon in his cla.s.sic tale From the Earth to the Moon, one can compute that the ultimate velocity that one can attain with gunpowder is only a fraction of the velocity necessary to send someone to the moon. Rail guns, however, are not limited by the speed of shock waves.

But there are problems with the rail gun. It accelerates objects so fast that they usually flatten upon impact with the air. Payloads have been severely deformed in the process of being fired out of the barrel of a rail gun because when the projectile hits the air it"s like hitting a wall of bricks. In addition, the huge acceleration of the payload along the rails is enough to deform them. The tracks have to be replaced regularly because of the damage caused by the projectile. Furthermore, the g-forces on an astronaut would be enough to kill him, easily crushing all the bones in his body.

One proposal is to install a rail gun on the moon. Outside the Earth"s atmosphere, a rail gun"s projectile could speed effortlessly through the vacuum of outer s.p.a.ce. But even then the enormous accelerations generated by a rail gun might damage the payload. Rail guns in some sense are the opposite of laser sails, which build up their ultimate speed gently over a long period of time. Rail guns are limited because they pack so much energy into such a small s.p.a.ce.

Rail guns that can fire objects to nearby stars would be quite expensive. In one proposal the rail gun would be built in outer s.p.a.ce, extending two-thirds of the distance from Earth to the sun. It would store solar energy from the sun and then abruptly discharge that energy into the rail gun, sending a 10-ton payload at one-third the speed of light, with an acceleration of 5000 g"s. Not surprisingly, only the st.u.r.diest robotic payloads would be able to survive such huge accelerations.

THE DANGERS OF s.p.a.cE TRAVEL.

Of course, s.p.a.ce travel is no Sunday picnic. Enormous dangers await manned flights traveling to Mars, or beyond. Life on Earth has been sheltered for millions of years: The planet"s ozone layer protects the Earth from ultraviolet rays, its magnetic field protects against solar flares and cosmic rays, and its thick atmosphere protects against meteors that burn up on entry. We take for granted the mild temperatures and air pressures found on the Earth. But in deep s.p.a.ce, we must face the reality that most of the universe is in turmoil, with lethal radiation belts and swarms of deadly meteors.

The first problem to solve in extended s.p.a.ce travel is that of weightlessness. Long-term studies of weightlessness by the Russians have shown that the body loses precious minerals and chemicals in s.p.a.ce much faster than expected. Even with a rigorous exercise program, after a year on the s.p.a.ce station, the bones and muscles of Russian cosmonauts are so atrophied that they can barely crawl like babies when they first return to Earth. Muscle atrophy, deterioration of the skeletal system, lower production of red blood cells, lower immune response, and a reduced functioning of the cardiovascular system seem to be the inevitable consequences of prolonged weightlessness in s.p.a.ce.

Missions to Mars, which may take several months to a year, will push the very limits of the endurance of our astronauts. For long-term missions to the nearby stars, this problem could be fatal. The starships of the future may have to spin, creating an artificial gravity via centrifugal forces in order to sustain human life. This adjustment would greatly increase the cost and complexity of future s.p.a.ceships.

Second, the presence of micrometeorites in s.p.a.ce traveling at many tens of thousands of miles per hour may require that s.p.a.ceships be equipped with extra shielding. Close examination of the hull of the s.p.a.ce Shuttle has revealed evidence of several tiny but potentially deadly impacts from tiny meteorites. In the future, s.p.a.ceships may have to contain a special doubly reinforced chamber for the crew.

Radiation levels in deep s.p.a.ce are much higher than previously thought. During the eleven-year sunspot cycle, for example, solar flares can send enormous quant.i.ties of deadly plasma racing toward Earth. In the past, this phenomenon has forced the astronauts on the s.p.a.ce station to seek special protection against the potentially lethal barrage of subatomic particles. s.p.a.ce walks during such solar eruptions would be fatal. (Even taking a simple transatlantic trip from L.A. to New York, for example, exposes us to about a millirem of radiation per hour of flight. Over the course of our trip we are exposed to almost a dental X-ray of radiation.) In deep s.p.a.ce, where the atmosphere and magnetic field of the Earth no longer protect us, radiation exposure could be a serious problem.

SUSPENDED ANIMATION.

One consistent criticism of the rocket designs I have presented so far is that even if we could build such starships, it would take decades to centuries to reach nearby stars. Such a mission would need to involve a multigenerational crew whose descendants would arrive at the final destination.

One solution, proposed in such movies as Alien and Planet of the Apes, is for s.p.a.ce travelers to undergo suspended animation; that is, their body temperature would be carefully lowered until bodily functions almost cease. Animals that hibernate do this every year during the winter. Certain fish and frogs can be frozen solid in a block of ice and yet thaw out when the temperature rises.

Biologists who have studied this curious phenomenon believe that these animals have the ability to create a natural "antifreeze" that lowers the freezing point of water. This natural antifreeze consists of certain proteins in fish, and glucose in frogs. By flooding their blood with these proteins, fish can survive in the Arctic at about-2C. Frogs have evolved the ability to maintain high glucose levels, thereby preventing the formation of ice crystals. Although their bodies might be frozen solid on the outside, they are not frozen on the inside, allowing their bodily organs to continue to operate, albeit at a reduced rate.

There are problems with adapting this ability to mammals, however. When human tissue is frozen, ice crystals begin to form inside the cells. As these ice crystals grow, they can penetrate and destroy cell walls. (Celebrities who want to have their heads and bodies frozen in liquid nitrogen after death may want to think twice.) Nevertheless, there has been recent progress in limited suspended animation in mammals that do not naturally hibernate, such as mice and dogs. In 2005 scientists at the University of Pittsburgh were able to bring dogs back to life after their blood had been drained and replaced by a special ice-cold solution. Clinically dead for three hours, the dogs were brought back to life after their hearts were restarted. (Although most of the dogs were healthy after this procedure, a few suffered some brain damage.) That same year scientists were able to place mice in a chamber containing hydrogen sulfide and successfully reduce their body temperature to 13C for six hours. The metabolism rate of the mice dropped by a factor of ten. In 2006 doctors at Ma.s.sachusetts General Hospital in Boston placed pigs and mice in a state of suspended animation using hydrogen sulfide.

In the future such procedures may be lifesaving for people involved in severe accidents or who suffer heart attacks during which every second counts. Suspended animation might allow doctors to "freeze time" until patients can be treated. But it could be decades or more before such techniques can be applied to human astronauts, who may need to be in suspended animation for centuries.

NANOSHIPS.

There are several other ways in which we might be able to reach the stars via more advanced, unproven technologies that border on science fiction. One promising proposal is to use unmanned probes based on nanotechnology. Throughout this discussion I have a.s.sumed that starships need to be monstrous devices consuming vast amounts of energy, capable of taking a large crew of human beings to the stars, similar to the starship Enterprise on Star Trek.

But a more likely avenue might be initially to send miniature unmanned probes to the distant stars at near the speed of light. As we mentioned earlier, in the future, with nanotechnology, it should be possible to create tiny s.p.a.cecraft that exploit the power of atomic and molecular-sized machines. For example, ions, because they are so light, can easily be accelerated to near the speed of light with ordinary voltages found in the laboratory. Instead of requiring huge booster rockets, they might be sent into s.p.a.ce at near the speed of light using powerful electromagnetic fields. This means that if a nan.o.bot were ionized and placed within an electric field, it could effortlessly be boosted to near light speed. The nan.o.bot would then coast its way to the stars, since there is no friction in s.p.a.ce. In this way, many of the problems plaguing large starships are immediately solved. Unmanned intelligent nan.o.bot s.p.a.ceships might be able to reach nearby star systems at a mere fraction of the cost of building and launching a huge starship carrying a human crew.

Such nanoships could be used to reach nearby stars or, as Gerald Nordley, a retired Air Force astronautical engineer, has suggested, to push against a solar sail in order to propel it through s.p.a.ce. Nordley says, "With a constellation of pinhead-sized s.p.a.cecraft flying in formation and communicating with themselves, you could practically push them with a flashlight."

But there are challenges with nano starships. They might be deflected by pa.s.sing electric and magnetic fields in outer s.p.a.ce. To counteract these forces, one would need to accelerate the nanoships to very high voltages on the Earth so they would not be easily deflected. Second, we might have to send a swarm of millions of these nan.o.bot starships to guarantee that a handful would actually make it to their destination. Sending a swarm of starships to explore the nearest stars might seem extravagant, but such starships would be cheap and could be ma.s.s-produced by the billions, so that only a tiny fraction of them would have to reach their target.

What might these nanoships look like? Dan Goldin, former head of NASA, envisioned a fleet of "c.o.ke-can sized" s.p.a.cecraft. Others have talked about starships the size of needles. The Pentagon has been looking into the possibility of developing "smart dust," dust-sized particles that have tiny sensors inside that can be sprayed over a battlefield to give commanders real-time information. In the future it is conceivable that "smart dust" might be sent to the nearby stars.

Dust-sized nan.o.bots would have their circuitry made by the same etching techniques used in the semiconductor industry, which can create components as small as 30 nm, or roughly 150 atoms across. These nan.o.bots could be launched from the moon by rail guns or even by particle accelerators, which regularly send subatomic particles to near light speed. These devices would be so cheap to make that millions of them could be launched into s.p.a.ce.

Once they reached a nearby star system, the nan.o.bots could land on a desolate moon. Because of the moon"s low gravity, a nan.o.bot would be able to land and take off with ease. And with a stable environment such as a moon would provide, it would make an ideal base of operations. The nan.o.bot could build a nanofactory, using the minerals found on the moon, to create a powerful radio station that could beam information back to Earth. Or the nanofactory could be designed to create millions of copies of itself to explore the solar system and venture off to other nearby stars, repeating the process. Because these ships would be robotic, there would be no need for a return voyage home once they had radioed back their information.

The nan.o.bot I"ve just described is sometimes called a von Neumann probe, named after the famed mathematician John von Neumann, who worked out the mathematics of self-replicating Turing machines. In principle, such self-replicating nan.o.bot s.p.a.ceships might be able to explore the entire galaxy, not just the nearby stars. Eventually there could be a sphere of trillions of these robots, multiplying exponentially as it grows in size, expanding at nearly the speed of light. The nan.o.bots inside this expanding sphere could colonize the entire galaxy within a few hundred thousand years.

One electrical engineer who takes the idea of nanoships very seriously is Brian Gilchrist of the University of Michigan. He recently received a $500,000 grant from NASA"s Inst.i.tute for Advanced Concepts to explore the idea of building nanoships with engines no bigger than a bacterium. He envisions using the same etching technology used in the semiconductor industry to create a fleet of several million nanoships that will propel themselves by ejecting tiny nanoparticles that are only tens of nanometers across. These nanoparticles would be energized by pa.s.sing through an electric field, just as in an ion engine. Since each nanoparticle weighs thousands of times more than an ion, the engines would pack much more thrust than a typical ion engine. Thus the nanoship engines would have the same advantages as an ion engine, except they would have much more thrust. Gilchrist has already begun etching some of the parts for these nanoships. So far he can pack 10,000 individual thrusters on a single silicon chip that measures 1 centimeter across. Initially he envisions sending his fleet of nanoships throughout the solar system to test their efficiency. But eventually these nanoships might be part of the first fleet to reach the stars.

Gilchrist"s proposal is one of several futuristic proposals being considered by NASA. After several decades of inactivity, NASA has recently given some serious thought to various proposals for interstellar travel-proposals that range from the credible to the fantastic. Since the early 1990s NASA has hosted the annual Advanced s.p.a.ce Propulsion Research Workshop, during which these technologies have been picked apart by teams of serious engineers and physicists. Even more ambitious is the Breakthrough Propulsion Physics program, which has explored the mysterious world of quantum physics in relation to interstellar travel. Although there is no consensus, much of their activity has focused on the front-runners: the laser sail and various versions of fusion rockets.

Given the slow but steady advances in s.p.a.ceship design, it is reasonable to a.s.sume that the first unmanned probe of some sort might be sent to the nearby stars perhaps later in this century or early in the next century, making it a Cla.s.s I impossibility.

But perhaps the most powerful design for a starship involves the use of antimatter. Although it sounds like science fiction, antimatter has already been created on the Earth, and may one day provide the most promising design yet for a workable manned starship.

10: ANTIMATTER AND ANTI-UNIVERSES.

The most exciting phrase to hear in science, the one that heralds new discoveries, is not "Eureka" (I found it!) but "That"s funny..."

-ISAAC ASIMOV.

If the man doesn"t believe as we do, we say he is a crank, and that settles it. I mean, it does nowadays, because now we can"t burn him.