Throughout the first half of the 20th century, the idea of using sunlight-driven sails for propulsion through the solar system and beyond was a recurring topic in both scientific literature and science fiction. Given the state of technology, fiction was perhaps the better label.
Soon after the demonstration of the laser in 1960, the realization that coherent laser light can form low-divergence beams led to a number of proposals for light sails propelled by laser photon momentum. Robert Forward’s not-so-modest 1984 proposal called for a space-based 100 GW laser array focused by a 100 km diameter Fresnel lens onto a 4 km diameter light sail and instrument package. The spacecraft would accelerate to 0.1 c in order to reach our neighboring star system, Alpha Centauri, in 40 years.
On 12 April, internet billionaire and former physicist Yuri Milner announced his plunking down of $100 million to seed research and development of a new, exceptionally ambitious laser-propelled light sail project: Breakthrough Starshot. The goal: reaching Alpha Centauri in 20 years.
What has changed since 1984? At least two things: major size and cost reductions in laser and instrument technologies enabled by Moore’s law, and the arrival on the scene of enlightened, monumentally wealthy internet titans. It’s admirable that those entrepreneurs want tech to aspire to more than liking, friending, and swiping. It’s also safe to say that Starshot is a tad more ambitious than online dating apps.
The task is to accelerate a 4 m × 4 m light sail and instrument package—all of 1 g—to a cruising velocity of 0.2 c. To produce the necessary 1013 J for a sail with 100% reflection, the project would require a 1-km-wide laser array beaming an average power of 100 GW for approximately two minutes. The light sails, each the cost of an iPhone, would be dealt out daily from a big Pez dispenser in Earth orbit and laser-accelerated to the target. After a year of daily shots, hundreds of probes will have been directed at Alpha Centauri, with a few possibly surviving to encounter it.
Yuri Milner’s Breakthrough Starshot project aims to use an array of ground-based lasers to propel lightweight probes to Alpha Centauri. (Image source: Breakthrough Initiatives.)
Let’s start with a rough estimate on laser cost, excluding electricity. Diode-pumped fiber lasers offer the highest average power bang for the buck, at an extremely optimistic off-the-shelf estimate of $100/W. That makes today’s lowball cost $10 trillion for 100 GW. Big ouch. Assume aggressive 90% savings through economies of scale (bringing the fiber laser below the cost of its diode pumps), and the cost could be decreased to $1 trillion. Add some Moore’s law mojo to cut that figure by another factor of 10: $100 billion. Still ouch. To put that in perspective, note that the Apollo moonshot program cost $150 billion in 2016 dollars.
The laser is the highest capital cost item, but electricity is another burden. The Starshot laser array would require approximately $1 million per laser shot at the US industrial rate of $0.10/kWh, assuming 50% wall-plug efficiency and a dedicated 100 MW power plant running 24/7.
More fundamentally, potential deal breakers exist that even endless boatloads of dollars may have trouble resolving. The main issue is the ability to quickly beam up the 1013 J to the light sail waiting in space for its star shot. Or equivalently, for how long can an array of lasers maintain a roughly constant average beam intensity on the sail?
A million 100-kW single-mode fiber lasers would be needed to supply 100 GW. (By the way, such high-power single-mode lasers don’t exist yet and present formidable technical difficulties.) If each of those aspirational lasers occupies 1 m2, then the laser array diameter would be about 1 km. To keep the full laser power trained on the 4 m × 4 m sail, the waist, or highest intensity location of the beam from the combined array, would need to move along with the sail as it accelerates. By controlling the relative phasing of the array’s contributing beams, the waist can be dynamically projected out to a maximum distance set by the array diameter. For those sizes of sail and array, the distance would be about 3 million km, for which the sail acceleration would be about 100 000 g to reach a speed of 0.2 c after about 100 seconds.
Even if, say, a graphene-based sail with a huge tensile strength could hold up to such g-forces and not be vaporized into space smoke by the laser, a major issue remains: The lasers must be coherently combined with precise phase control so that they don’t act like 1 million independent beams. How easy is it to keep 1 million lasers in phase at the rapidly accelerating sail after propagation through tens of kilometers of a windy, turbulent, aerosol-filled atmosphere?
Some say that problem is solvable with adaptive optics and feedback. The less optimistic view, to which I subscribe, says that no feedback system could possibly make the needed laser phase-front adjustments quickly or accurately enough. Rapid fluctuations in the intrinsic phases of the lasers would occur midflight, and air heated by the high-power beams would create additional turbulence. One thing is certain: Unless the apparently insurmountable problem of laser phasing is solved, a ground-based laser array will not work.
Despite the scale and complexity of a laser system on Earth, locating it in space—where the turbulence problem is solved—introduces an impossibly more complex set of issues beyond the wildly amplified costs and logistics. Simply put: A space-based 100 GW laser is a de facto superweapon, the costliest in the history of the world. (Cue Shirley Bassey.) Such a project would set off interminable national and international political wrangling.
There is a bright side to all this. Milner’s very generous seed grant—and the attendant publicity—is a timely and needed jump start for the imagination, just as the US and Russian space programs were for many of us as children. Given the off-the-charts technical challenges and costs to going interstellar, $100 million isn’t remotely enough. Maybe no amount is. Still, for investigating some of the basic enabling science (which is important in its own right), and for perhaps refocusing attention on what tech should be about, Milner’s effort is a commendable public service.
Howard Milchberg is a professor of physics and electrical and computer engineering at the University of Maryland, College Park.
