"YES yes," the lady said to me after a talk I gave on writing science fiction, "but what are you doing yourself in science?" I was tempted to say that unlike most writers, and to the eternal gratitude of my parents, I was holding down a full-time job.

But the humor would be missed. Like three others who got Ph.D.s at the University of California, San Diego—Vernor Vinge, David Brin and Kim Stanley Robinson—I went steadily downhill after graduation, becoming a published author. But unlike them, I still do science when I get the chance.

I get letters and e-mails asking what I do in science, so this column traces out my primary work over the last few years, both in theory and experiment.

I've spent a lot of time working on the Planetary Society's hopes to launch its Cosmos Sail in early Fall, 2003. This thirty-meter-wide aluminized mylar disk will be the first spacecraft driven solely by sunlight pressure. Deployed from a Russian launch vehicle, the sail will open its fifteen-meter vanes at 800 kilometers altitude and then rotate them to catch sunlight's subtle pressure. It will finally realize a 75-year-old idea.

The flight engineers hope to pump its orbit higher by turning the vanes full on to sunshine on one half the orbit, then rotate them ninety degrees to avoid braking on the second half. That may raise the orbit enough to avoid deorbiting it in two months. But after a month, the Sun will get some help.

A microwave beam from the Goldstone 100-meter antenna, largest in the Deep Space Network, will reflect from the sail. This will be the first known attempt to exert forces on a spacecraft from the ground. The idea of doing this occurred to me when I was helping plan the mission. I admit it, I was inspired by those old magazine covers showing beams moving big things in space.

Goldstone's steerable dish radiates up to half a megawatt, but because the sail will be beyond the focal range, the beam will hit it with only about 1700 Watt. Sunlight pressure will accelerate the sail with at most 10-4 of a gravity, and the beam will manage roughly 10-7. Still, onboard accelerometers can measure this as an in-principle demonstration of beamed power in space. The goal is to illustrate future possibilities, not usefully move the sail. The craft will send the data back to the Planetary Society's downlink center.

The beam-driven sail idea dates from 1966, when pushing light mission packages with lasers seemed a natural outgrowth of solar sailing. When illuminating the Cosmos sail first occurred to my brother James and me, we studied using a large Air Force laser for this experiment. But the laser costs a million dollars a minute to fire, whereas Goldstone's beam costs only a few hundred dollars, and NASA is picking up the cost.

Accelerating a sail depends only on the power, not the frequency of the beam. Microwave transmitters have been under development much longer than lasers; they are far more efficient and much cheaper to build. Their disadvantage is that they must have much larger antennas for the same focusing ability, but that does not matter in this case. Also, microwaves do not damage sail materials as lasers can and do not refract while passing through air.

The point of this effort is to see what a truly twenty-first-century spacecraft might look like. I've done a lot of calculations and experiments in my lab at UC Irvine as a consultant to NASA.

Whatever the source of the beam (power supply plus antenna, the "beamer"), the basic ability to move energy and force through space without moving mass is key to a new sort of spacecraft. The expensive part of this utility is the beamer, which stays on the ground where we can fix it, improve it, and then project energy anywhere within its range. Because they are low-mass (a few hundred kilograms), sails of aluminized mylar (or even better, carbon fiber that can withstand high temperatures) can be accelerated to high velocities, perhaps making fast missions beyond the solar system possible.

Like the nineteenth-century railroads, once the track is laid, the train itself is a small added expense. Compared with rockets, sails are very cheap, once the beamer is built. Just as railroads opened the American West, a beamer on Earth—or for better focusing ability, in orbit—could open up the frontiers of our solar system and beyond.

The spacecraft would be light and fairly cheap, so many could be sent at low per-shot cost. The low mass of sails could allow launch from Earth-based or orbiting microwave transmitters, imparting high velocities.

Interplanetary spacecraft must fight their way out of the Earth's gravitational well, but the neglected virtue of this is that sailcraft that have not escaped Earth's clasp must return on an elliptical orbit. A sail will repeatedly revisit a beamer in orbit, climbing to higher altitudes as the beamer's impulses add each velocity increment. After hundreds of orbital raisings, the sail departs into interplanetary space, where sunlight can push it farther.

Other applications include fast missions to Mars, if an eventual manned expedition needs low-mass replacement parts or medical supplies. A sail could decelerate in the Martian atmosphere, then descend by parachute.

To study such ideas, a team including me and my brother Jim has actually "flown" sails at JPL and UC Irvine. We did experiments with both tissue-thin aluminum sails and with small sails a few inches across made of pure carbon fibers.

Ten times thinner than a human hair, these micro-fiber mats can endure very high temperatures. They weigh in the range of ten grams per square meter—lighter than tissue paper, and competitive with the very lightest aluminized mylar sails. They are intrinsically stiff as well, and can remember their shape after being rolled or folded, as deployment tests have demonstrated.

Carbon sails could dive to very near the Sun and withstand heating far beyond possible with current spacecraft, up to 2000°K. Theoretically, this opens up missions for sails accelerated by ultrastrong sunlight to velocities in the range of 100 km/sec, for fast missions beyond Pluto.

It's a cute idea—but could we show it? We put the sails in a chamber the size of a Volkswagen, pumped out the air, and hit it with microwaves. The very first try, the sail lit up (hot!)—then flew up and smacked onto the chamber ceiling. Cheers.

Repeatedly we showed lifting and upward flight of ultralight sails. The carbon-carbon microtruss material easily survived several gees acceleration. To propel the material, we sent a ten-kW microwave beam into a vacuum chamber. At microwave powers a hundred thousand times sunlight, the sails reached 2000°Kelvin (from microwave absorption) and survived. Carbon is one of the few materials that can take such temperatures and survive as a structure. This capability of carbon rules out most materials for hot, high acceleration missions. For example, present spacecraft would melt away at 1000°Kelvin.

Still, a mystery arose. Data analysis and comparison with candidate acceleration mechanisms showed that the beam's purely electromagnetic pressure accounted for only three to thirty percent of the observed acceleration, so another cause was acting.

This led to another new idea. I remember reviewing the data and suddenly thinking, "This mysterious effect messed up a nice experiment, but maybe the universe is trying to tell us something. This is better than the idea of pushing sails with just light pressure. But what?"

Back to the data.

Analyzing the gases blown off the sails with the beam on, we found that the main thrust came from molecules embedded in the carbon fibers during manufacture. This is called sublimation or desorption, and the higher the temperature, the more thrust results. We believe that the main lift in our experiments came from carbon monoxide being liberated from the carbon fibers, at temperatures above 2300°K.

The thought immediately came: This might be a useful propulsion mechanism—a wedding of the solar sail idea with classic rocket engineering. Flat sails make poor rockets because there is no nozzle. On the other hand, they carry no engine.

How effective is desorption relative to the photon reflection for which sails are designed? The ratio of accelerations is greater than ten, and can be as high as ten thousand at high temperatures. For example, for molecular hydrogen, the ratio is 10,000 for temperatures of 1000°Kelvin. This means that a beam source can exceed acceleration by sunlight if it illuminates the sail for only a small fraction of the sail's orbit time around the Earth. Such a large multiplier is the essence of the assisted beam-driven method.

Our calculations show that this could shorten the escape time from Earth's gravity well to weeks, compared with years for solar sails. The sail returns to near the beam source on each loop of a steepening ellipse. Gravity is the enemy, but at least it does bring the sail back to a beamer—say, one sitting in a circular orbit, awaiting—on an obliging ellipse.This would be a unique advantage to beam-driven sails, enabling repeated high accelerations and course corrections.

Plausible scenarios using about 100 MW microwave beam powers allow fast beam-plus-solar sailing missions to the outer solar system. This in turn opens missions using the close approaches to the Sun.

A voyage beyond Pluto could begin with a carbon sail's deployment in Low Earth Orbit by conventional rocket.

An orbiting beamer then launches the sail from nearby with a microwave beam in orbit. Once free of Earth, it can use sunlight to navigate inward to near the Sun.

I called this craft the Sundiver. The term is old—I gave it to David Brin when he first came to see me, back when he was struggling with his first novel. (As he now recounts, I asked him how his craft that literally plunges into the Sun could survive. He answered that he would throw in some jargon, techtalk, whatever. I disdainfully replied, "Oh—magic." So David went home and found a physically possible way to do it, confounding me.)

Consider the sundiving sail. Approaching the Sun turned edge-on (to prevent the increasing flux of sunlight from pushing against its fall), the carbon sail heats up. At closest approach, the craft could turn to absorb the full glare of the intense Sun, gaining a high velocity as it accelerates strongly, under desorption. It exhausts the store of molecules lodged in its fibers, losing mass while gaining velocity. It then sails away as a conventional, reflecting solar sail. Its final speed could be high enough to take it beyond Pluto within five years. There it could do a high velocity mapping of the outer solar system, the heliopause and beyond, to the interstellar medium—the precursor to true interstellar exploration.

Such maneuvers demand a lot of sail acrobatics. The worst problem, as we discovered in experiment, recalled a classic stunt. Chinese performers can balance plates on the ends of sticks by spinning them; without spin, they fall off. A sail riding a beam is in the same fix. Spinning helps a lot. But how to spin it up, and keep adjusting spin for the whole ride? Could we use the beam to do this?

Back to the notebooks.

In experiments at JPL and UC Irvine we used circularly polarized beams to make carbon sails spin by absorption of the beam. The angular momentum in the beam simply gets deposited in the sail. Microwave powers of 100 watts—the power of a light bulb!—spun carbon cones a few cm across up to a cycle/second.

Somewhat surprisingly, even good electrical conductors like aluminum can be spun if they are not cylindrically symmetric. This is a geometric effect from interference of the waves in the beam when they reflect from the sail.

Classic disk sails won't spin, but introducing cuts or struts or making them otherwise nonsymmetric lets them spin readily. Sometimes this geometric approach proves more effective than through material absorption, as with carbon. All this was new to electrodynamics, a field 150 years old, but still rich in new phenomena.

As a mechanism to unfurl sails in space, electrodynamic spinning allows the beamer both to push and to spin with the same beam. Here, too, lasers fail. Since the spinning effect depends upon the wavelength of the electromagnetic beam, the far shorter wavelengths of lasers cannot spin sails.

With spin, stability and control during beam-riding become easier. Even if the beam is steady, a sail can wander off the beam if its shape becomes deformed, or if it does not have enough spin to keep its angular momentum aligned with the beam direction in the face of disturbances.

Generally, sails without structural elements cannot be flown if they are convex toward the beam, as the beam pressure would make them collapse. On the other hand, the beam pressure keeps concave shapes in tension, so they arise naturally while beam riding. They will resist sidewise motions if the beam moves off center, since a responding net sideways force restores the sail to its position.

Therefore, we concentrated on a conical shape for the sail and studied its dynamics in numerical simulations. Experimental data showed that the beam-riding effect does in fact occur. With microwave powers of a few hundred Watts we could hold an otherwise unstable sail steady, if the focused beam power falls off fairly quickly with angle from the central axis.

We are now studying how active feedback can stabilize such sails, with a team at the University of New Mexico. Those Chinese spinning-plate acts knew a lot we're just discovering. So far, the only sail shape that is stable, riding the beam, is shaped like a shallow Chinese hat—not a disk! Who knew?

These ideas and experiments interlock with another older idea: transmitting solar energy collected by platforms in orbit down to Earthly consumers. Receivers on the ground would collect the microwave beams and turn them into electrical power.

Such Space Solar Power, or SSP, intersects these sail ideas well. A beamer would be the workaday SSP array, but then could be used for only minutes at a time to push a sail as it came around again in its lengthening, elliptical orbit. Uniting domestic energy technology with deep space exploration answers the critics who say NASA's explorations yield little benefit.

More exotic approaches beckon in future. Advanced "smart sails" could have electronic circuits dispersed in the sail area. The circuit elements would not be wires but rather the carbon fibers themselves. Carbon carries electrical current, and with future developments could carry out on-board computing. Uniting such functions means that the same mass in carbon both absorbs momentum, electrical energy (charging its batteries) and even broadcasts back to Earth on command, using the type of phased array circuitry that the Deep Space Network employs every day. The sail becomes its own antenna.

All these ideas beckon at our horizons. To make the solar system ours, we must envision using propulsion methods beyond those of the chemical rockets developed more than half a century ago. The railroad was a utility that still does yeoman work today, though it gave way to the auto and the airplane.

Sending energy and momentum through space faces limits in the focusing ability of antennas and the properties of ultra-light materials. Before we see spacecraft handled at a distance purely electromagnetically, in true hands-off style, we will have to use bold, fresh thinking.

So what does twenty-first century space flight look like? Plenty of beam-assisted sails zooming around the solar system and beyond, each one fairly cheap and thus expendable. No more precious craft like Cassini (due at long last to reach Saturn in July 2004, a project that began in the late 1970s) whose loss would mean a billion bucks down the drain.

Nuclear rockets to move people and supplies. Beam-driven sails to give fast, pony express backup to manned expeditions on Mars or the asteroids. Break a five-gram widget? Ask for one pronto on the sail express. The Space Solar Power utility takes a few minutes of its time—usually it's exporting gigaWatts of power to power grids down on the Earth—to push the sails out into interplanetary space. The sails are a sideline to the real business of powering the ever-power-hungry multitudes below.

But of course we have a long way to go to make this happen. Basic physics—my line of work—must be followed up by real engineers who find out how to fly the tricky, light craft. Building the beamer in low Earth orbit will be pricey, maybe several hundred million dollars—but like railroad track, it would pay for itself over time.

All this hinges on how much we want to explore, to venture, and perhaps to profit in space. Alas, that's politics—not my area.

I prefer to stay in the lab, pushing my pencil in calculations. It's closer to the future, and more fun.

copyright © 2003 Abbenford Associates

Gregory Benford is a professor of physics at the University of California, Irvine. Comments on this column welcome at gbenford@uci.edu, or Physics Dept., Univ. Calif., Irvine, CA 92717