Breakthrough Starshot: Mission to Alpha Centauri

by Paul Gilster on April 12, 2016

Here on Centauri Dreams we often discuss interstellar flight in a long-term context. Will humans ever travel to another star? I’ve stated my view that if this happens, it will probably take several hundred years before we develop the necessary energy resources to make such a mission fit within the constraints of the world’s economy. This, of course, assumes the necessary technological development along the way — not only in propulsion but in closed-loop life support — to make such a mission scientifically plausible. I get a lot of pushback on that because nobody wants to wait that long. But overall, I’m an optimist. I think it will happen.

Let’s attack the question from another direction, though, and leave human passengers for a later date, as Yuri Milner’s Breakthrough Initiatives, aided by Stephen Hawking, is doing today in a New York news conference. What if we talk about unmanned missions? What if, in fact, the question is: How soon can we put a scientific payload past another star? Let’s not worry about decelerating — this will be a flyby mission. Let’s build it as soon as possible using every breakthrough technology we have at our disposal. How long would it take for that mission to be developed and flown?


Milner, a philanthropist and investor who was an early backer of Facebook, Twitter, Spotify and numerous Chinese tech companies, tells me his goal is to ‘give back to physics’ in developing just such a mission. Part of that giving back is the $100 million he has already put forward to support SETI, a ten-year project that will produce more telescope time for SETI than any other. Milner is also the founder of the Breakthrough Prize, issuing awards in physics, life sciences and mathematics. But in many respects this third Breakthrough Initiative is the most daring of all.

Time for the Stars

Breakthrough Starshot is an instrumented flyby of Alpha Centauri with an exceedingly short time-frame, assuming research and development proceed apace. Milner is putting $100 million into the mission concept, an amount that dwarfs what any individual, corporation or government has ever put into interstellar research. A discipline that has largely been the domain of specialist conferences — and in the scheme of things, not many of those — now moves into a research enterprise with serious backing.

Could an Alpha Centauri flyby mission be developed and launched within a single generation? I think it’s quite a stretch, but it’s the best-case scenario Milner mentioned in a phone conversation over the weekend. He’s enough of a realist (with a first-rate physics background) to know that the challenges are immense. Even so, he sees no deal-breakers.

Let’s walk through the case and see why he finds reason for optimism. “There are major advances that we can now turn to as we develop this proof of concept,” Milner says. “Twenty years ago, none of these things would have been available to far-thinking scientists like Robert Forward. But now we can put them to use and test their possibilities.”

If you’re thinking of an interstellar mission in the near-term, there is really only one choice of propulsion: The beamed sail. Sails have the advantage of known physics, laboratory experiment and actual deployment in space. We could talk about fusion for some indefinite point in the future, but at present, we don’t know how to do fusion even in massive installations on Earth, much less in the tight confines of a spacecraft engine. Interstellar ramjets are a far-future unknown — they may act more effectively as braking devices than engines, according to recent research. Antimatter is nowhere near readiness for propulsion, either in production methods or storage. Chemical rockets fall victim to the mass/ratio problem and are useless for fast interstellar journeys.

That leaves us with sails carrying very small payloads. To cross the 4.37 light years to the Centauri A and B system, Breakthrough Starshot proposes small spacecraft, taking advantage of advances in nanotechnology to reduce payload size. Think Moore’s Law and the reductions in size and cost that have accompanied the vast increases in micro-chip power. “Moore’s Law,” says Milner, “tells us that now is the time.”

StarChip is the Breakthrough Initiatives’ name for a payload measured not in kilograms but grams, a wafer that carries everything you would expect in a fully functional probe. ‘What was once a 300 gram instrument is is now available at three grams,” Milner continues. “What was 100 grams is now 0.5 grams. This is the trend we are riding.”

The StarChip payload includes cameras, power supply, communications equipment, navigation capabilities and photon thrusters. And it would be thrown across the interstellar gulf at 20 percent of the speed of light by a sail that is itself a miniaturized version of the sails Robert Forward used to discuss. Forget the thousand-kilometer sail (much less the continent-sized sails of the science fiction dreamer Cordwainer Smith). Milner’s team believes we can now talk in terms of a laser-driven lightsail that is no more than 4 meters across. This is actually smaller than the first deployed sail craft, the Japanese IKAROS, which boasts a sail measuring 14 meters to the side.

Advances in metamaterials and additional research should be able to produce, Milner believes, a 4 meter sail whose own weight is tallied in grams, and whose materials allow fabrication at a thickness of a few hundred atoms. A sail that small makes its own statement: Clearly, it’s not going to be under the beam for long, which means we need to focus a great deal of light on it for a very brief time. Lasers are another technology that benefits from rising power and falling cost. The trick here will be to create ‘phased arrays’ of lasers that can scale up to the 100 gigawatt level. A phased array involves not one but a group of emitters whose effective radiation pattern is reinforced in the desired direction by adjusting the phase of the signals feeding the antennae.

This is classic Bob Forward thinking rotated according to the symmetries of our new era. Milner aims for a beamer technology that is modular and scalable. And it fits into a larger infrastructure. Breakthrough Initiatives talks about bringing a ‘Silicon Valley approach’ to the problem of interstellar flight. Build a StarChip that can eventually be mass-produced at no more than the cost of an iPhone. For the Alpha Centauri mission, whenever it flies, is itself a proof of concept that could lead to multiple destinations. And if the cost can be driven as low as Milner believes, then we can think in terms of redundancy, with StarChips sent in large numbers to return a full characterization of any destination system. Assemble the light beamer and, as the technology matures, the cost of each launch falls.

These are ideas that are at once familiar but also exotic, for while Forward talked about enormous power stations in close solar orbit to power up his banks of lasers (and a huge Fresnel lens in the outer system to focus the beam), Milner thinks we can build a ground-based beamer at kilometer scale right here on Earth. I was startled at the idea — surely efficiency favors a space-based installation — but Milner’s point is that he thinks we can begin to launch interstellar craft before we have the technology to build the kind of power station Forward envisioned. If you’re serious about a launch within a few decades (again, it’s a best case scenario, and a dramatic one), then you build an Earth-based beamer and use adaptive optics to cancel out atmospheric effects.


Image: A wide-field view obtained with an Hasselblad 2000 FC camera by Claus Madsen (ESO), of a region around the Southern Cross, seen in the right of the image (Kodak Ektachrome 200, 70 min exposure time). Alpha Centauri is the bright yellowish star seen at the middle left, one of the “Pointers” to the star at the top of the Southern Cross. Although it appears here as a single ‘star,’ it is actually comprised of the G-class Centauri A, K-class Centauri B, and the M-dwarf Proxima Centauri. Credit: ESO/Claus Madsen. Original here.

All this will be subject to tightly focused research, which is what the $100 million is for, but what Milner hopes to see are nano craft delivered to orbit and then boosted on their way with a 30 minute laser ‘burn’ that, reaching 60,000 g’s, drives the sail to 20 percent of the speed of light. That makes for roughly a twenty year crossing to Alpha Centauri. With a craft this small, data return is highly problematic, and in fact I think it’s one of the biggest unanswered questions Breakthrough Starshot will have to face (well, this and the challenge of interstellar dust, and key questions related to sail design and the sail’s ability to stay on thee beam during acceleration). The sail is itself the antenna on a craft of this design, and Jim Benford told me in conversation that it will have to be shaped to one-micron precision. Even so, powering up the system to send imagery and data to Earth is going to be tricky. It will be fascinating to see what kind of solutions emerge as this research gets underway, and what alternative methods may be suggested.

Even so, and granting the cost reductions digital technology makes possible, Breakthrough Starshot embarks upon a multi-year research and engineering phase that will focus on building a mission infrastructure. Creating the actual mission will demand a budget comparable to the largest scientific experiments of our time. These are no small aspirations, but what drives them is something that interstellar studies have never had at their disposal: A dedicated, enthusiastic, well-funded effort with the participation of major scientists.

“We have an advisory board of twenty, including Freeman Dyson and other top scientists,” Milner added. “$100 million will be spent in coming years as we look toward concept verification. Multiple grants should flow from this, research and experiments. We need to complete the initial study and see if building a prototype, perhaps at a scale of 1/100, is then the next step.”

At the very least, we can expect the research behind this project to spin off numerous useful technologies, all of which should be applicable not only to star missions but to in-system exploration, along with, potentially, a kilometer-scale beamer that can double as a large telescope for astronomical observations. And while I doubt we can look at interstellar missions within the next few decades (I am open to being convinced otherwise), I believe that the timing for a fast flyby of Alpha Centauri will be considerably advanced by this work.

There is much to be said about all aspects of the Breakthrough Starshot concept, and as you would imagine, I’ll be covering this closely, beginning with a trip later this week to the Breakthrough Initiatives meeting in California. That meeting will have a large SETI component growing out of Milner’s prior commitment of another $100 million, which is already being translated into active observations at the Green Bank observatory in West Virginia. But as you can imagine, the Alpha Centauri mission will be under discussion as well as the research effort begins to be assembled. What spins out of this will keep us talking for a long time to come.


Thomas Goodey April 14, 2016 at 0:33

I suppose the necessary energy would be stored in a huge array of some type of supercapacitors, unspecified.

The waste heat would not be a problem. It could easily be dissipated. If you can build the lasers, you can certainly build a water-cooling system.

However the waste heat at the probe end is, in my opinion, an insoluble problem. If there is even one millionth part of reflection inefficiency, the probe will be immediately vaporized.

Micky Badgero April 14, 2016 at 9:55

Lubin addresses the probe heating in some of the details of the project. The nice thing about this project is that there are details. And $100M funding to look into problems like this.

One solution in his paper ( is to put the wafer edge on to the beam. Another would be to make the probe long and thin, instead of a round wafer to reduce the cross-section to the beam. This would also help with erosion against interstellar gas and dust during the trip (which was covered better by Project Daedalus than by Lubin). Another solution would be to reduce the beam power and accelerate at a modest pace instead of 60,000 g’s.

For me, power and the ‘supercapacitors’ are a bigger problem. We are nowhere close to making a gram-sized transmitter that can transmit from Mars, let alone Alpha Centauri.

Michael April 16, 2016 at 10:13

‘For me, power and the ‘supercapacitors’ are a bigger problem. We are nowhere close to making a gram-sized transmitter that can transmit from Mars, let alone Alpha Centauri.’

We can easily pick up 100 W signals, that’s only 100 time a mobile phone power output. As the craft moves deeper into space we need only build bigger receivers and wider baselines and we have large amounts of space and power to do it over time. Either we will use the gravitational focus points or relays to transmit and receive information. I suspect we will use the solar focus points for long range transmission between the stars and then relays to these focus points to craft and Earth.

Micky Badgero April 16, 2016 at 15:26

100W from Mars or from Alpha Centauri?

Michael April 17, 2016 at 2:48

From Mars, we would need seriously large receivers to pick up 100 W from AC!

Javier Garcia April 14, 2016 at 7:10

Noboy knows what will be the life expectancy in 50 years, but sure it will be higher than today, and even higher i you have plenty of money to pay for special treatments.

Javier Garcia April 14, 2016 at 7:20

Agreed. In fact I was really shocked when I saw this information among the head ones on the night tv news.

There will always be poor and rich. I am only remarking possible controversy.

Javier Garcia April 14, 2016 at 7:26

While “targeting the moon” is a good way of “flying high” with our dreams, and I think that is something good, an even necessary for the human being, I also agree that having more reachable targets in mind, better if there is a posibility to have some kind of economic return on them (mining, farmacy, etc…) woul be the best way to ensure the long term surviving of the space dream.

Michael April 14, 2016 at 9:50

Just recovering today from a ‘Mosh Pit’ experience with the ‘Overkill’ band fans last night in London…lets see broken glasses, dislocated finger, lost boot but it was thank fully returned and numerous dents to the body.

The ‘Mosh Pit’ experience has much in common with a mugging, it is just a lot more enjoyable..!

So whilst recovering I have had time to think about a fission implosion drive, this is where a thin shell of fissile material is imploded via laser/ions to high density to cause a small nuclear explosion. If we were to at the same time just when the collapsed shell reaches maximum density use a linac to generate a neutron stream it would be a lot more efficient. We can control the neutron beam focus down to almost the atomic level by using a nanotube focuser forest and have sub micron second generation times. Fissionable materials are 99% ready to blow up and are remarkable dense fuels, a craft built around this technology would be remarkably compact and a benefit is that all of the technologies are matured. If we were to use this type of machine to move through another star system we could, as I have mentioned before, drop small magsails into the exhaust steam to slow them down to enter into orbit or at least give us a longer viewing time.

meganixer April 14, 2016 at 16:44

now that you’ve come down from that high bounce mosh-pit recovery, do u still think this is the way.

there’s only one thing that really interests me about this project – further confirmation of general relativity with the high 0.2c speeds, time dilation, length contraction,

Michael April 15, 2016 at 2:31

All vital organs are back in their normal places now, and YES to your question, fission implosion drives are workable and very powerful.

The unfortunate problem with the sail concept is stopping at the SGFP for a good look, high velocities through this extended region will have a blurring effect. A powered drive will allow us to stop and view other objects near by as well and very deep into space.

Michael April 15, 2016 at 9:19

‘there’s only one thing that really interests me about this project – further confirmation of general relativity with the high 0.2c speeds, time dilation, length contraction,’

This has all been proven already in particle accelerators to umteenth sigma.

D Desai April 14, 2016 at 10:44

While I am completely in support of the concept, I wonder if a good (and more feasible in the near term) intermediate step would be to use the Starshot laser array and nanocraft to travel to the Sun’s gravitational lens focal point. This would involve a journey of about 600 AU or 17 days. And yes, there is something amazingly cool about writing ‘600 AU…17 days’! With the resolution of gravitational lensing (previously calculated as 2.8m/ly in another Centauri Dreams article) we might possibly get imagery comparable with a flyby. Additionally the communication time of about 80 hours, vs 4.37 years, would mean that we could have the first high-quality survey of the Alpha Centauri system easily in our lifetime…in about 3 weeks after launch vs 24 years.

Kent Kemmish August 26, 2016 at 22:49

That’s interesting, but do we need to wait for massive lasers to do that? Even if it took five years to get to the Sun’s gravitational lens focal point, that surely beats 24 years… and presumably, if we sent an observatory there, we could look at a lot more than just Alpha Centauri et al..

As a naive astrophysics/space travel layperson, this seems like a much better/bigger bang for the buck idea than the StarShot proposal. What am I missing?

2.8 m resolution per light year??!! What would we have to send to take advantage of that? Sounds like we could get highly detailed maps of every exoplanet out to thousands of light years with that resolution!…?

John Mccarthy April 14, 2016 at 12:38

Fortunately, there is no effect on electronics transiting the Van Allen Belt. I do hear concern for astronauts traveling to Mars when they too must transit the Belt. If that is the case, how did the 1969 moon landing happen re transiting the Van Allen Belt?

Andrei April 14, 2016 at 20:12

Cool idea Yuri! The lasers can be built, I have no doubt about that – so far so good.

However, I cant imagine what material that would have to be used to survive both such an acceleration and the radiation of 20 years in space. Even benign particles, and hydrogen atoms will have a tremendous energy at .2 C

Adding such an tremendous energy to the sail with an area of only a few square meters. If it would need to reflect 99,99 and a long line of nines of the light hitting it. There’s no such material known. Else the energy will simply melt the sail in a split second – regardless of it’s size or shape. At the same time it will have to be extremely rigid so it will not fall apart from that 6000G punch. Again, there’s such materials known – perhaps if it is a solid cube. But delicate micromachines?

More serious is that the slightest misalignment of the beam will have the sail spinning wildly, and quite possibly disintegrate in the tiniest fraction of a second.

If Breakthrough Starshot is to be anything more than just a cool demonstration that we’ve developed to capacity to send sails at 20% of lightspeed we also need to build the infrastructure to recieve data from the probes. We certainly do not have the capability today, but perhaps such a network can be built to pick up such a weak signal over interstellar distances. But to actually get useful data of lets say several megabyte…that will be tricky.
The main problem lies in the far end. We don’t use nano satellites today since the transmitting power is to low for interplanetary probes. Yes, such might be used in upcoming missions, but those will still have to send their data to the main craft that relay the data to Earth. Compare with Philae that sent data to Rosetta which in turn relayed the data to Earth. There’s no counterpart to that here.

So transmitting any data back to Earth is a problem that I don’t think can be solved in such a small package. What wattage could the transmitter have? A few watts at best. That would require impossibly tolerances for a very narrow transmission both in frequency and how it is aimed at Earth for such small electronics. (It can be done, but again its a matter of physics, the smaller the circuits and the miniature dish, the less precise it will be.)
So the physics of nature require a rather large disk. The weight limitation exclude one such.

Lastly, a probe not larger than suggested here cannot carry a telescope to take images with any resolution to talk of, physics -again- or more specifically diffraction dictate that the lens have to be of a decent size, lets say the same as one amateur telescope or at least common binoculars.

Even if the sail could be aimed with an almost magical precision to get within range to take images at a shorter distance to make the project worth pursuing. No! You cant go closer, the probe travel to fast to be able to take any image at close range of anything!

There’s one kind of data that can be obtained with a small lens while still going very close to a planetary target – and that’s to take spectra.
But! That is something we hope to be able to do quite soon from within of our own solar system. So there’s no reason to send a “Breakthrough” sail for such science, it could be done anyway and for many more stars with instruments closer to home that could be used to study many more target stars.

Sure go ahead and build interstellar sails, a wonderful idea. But it cannot be spin doctored as adding to our knowledge of distant stars and planets that could be obtained in less costly and less time consuming ways.

Adrian Bankhead April 14, 2016 at 22:02

my apologies for this prosaic question, but is Breakthrough Discuss 2016 (described at invite only? I live close to Stanford, and I was hoping to drive down there this weekend to attend the April 15-16 lectures, but time and location information is not listed. If this event is invite only, are the lectures going to be recorded, or is it private by design?

Paul Gilster April 15, 2016 at 10:01

Adrian, sorry to say that it’s an invitation-only conference. I don’t know about how the sessions will be recorded, but I know the plan is for open access to all data and I assume that includes the proceedings. I suspect they’ll be going online at some point, and will keep readers advised.

Andrew Norman April 17, 2016 at 0:41

In his white paper on his Destar concept (from which the Breakthrough starshot idea originated) Lubin proposes using reflected light from the propulsion laser for long distance communication. From his paper:

“Communications – Another use of the DE-STAR system would be for long range interstellar communications to and from the spacecraft. This is a critical issue for long range interstellar probes in the future. Can we get high speed data back? DE-STAR to spacecraft data rate – Consider an optical link calculation with DE-STAR 4 which emits about 50 GW at 1.06 µm or about  29 3 10 γ/s , with a divergence half angle of 10  D  10 rad (22) At a distance of L = 1 ly (~1016 m) the spot size (diameter) is about Ds ~ 2∙106 m. For the case of the 100 kg robotic craft and with a 30 m diameter reflector this gives a spacecraft received photon rate of 3∙1029 x (30/2∙106)2 ~ 7∙1019 γ/s. If we assume it takes 40 photons per bit (which is very conservative) this yields data rate of about 2∙1018 bits/s, clearly an enormous rate. See figure. Spacecraft to DE-STAR data rate – Assume the spacecraft has a very modest 10 W transmitter on the spacecraft (an RTG for example) for an optical link at the same basic wavelength ~1.06 µm (slightly different to allow full duplex communications if needed) and that it uses the same 30 m reflector as for the photon drive but this time it uses it as the communications transmitter antenna (mirror). We do the same basic analysis as above. 10 W at 1.06 µm or about D   x  19 5 10 γ/s 8 3.5 10 rad , with a divergence half angle of At a distance of L = 1 ly (~1016 m) the spot size (diameter) is about Ds ~ 3.5∙108 m. For the case of the 100 kg robotic craft and with a 30 m diameter reflector transmitting BACK to a DE-STAR 4 which acts as the receiver this gives a received (by the DE-STAR) photon rate of 5∙1019 x (104 /3.5∙108)2 ~ 4∙1010 γ/s. If we assume it takes the same 40 photons per bit this yields a received (at Earth or wherever the DE-STAR system is located) data rate of about 1∙109 bits/s or 1 Gbps. At the nearest star (Proxima Centauri) at a distance of about 4 ly the data rate at Earth from the spacecraft is about 70 Mbps. Live streaming >HD video looks feasible all the way to our nearby interstellar neighbors. This is summarized in Figure xx. “

Andrew Norman April 17, 2016 at 0:56

Sorry- I misread this the first time. He is proposing using of the propulsion laser as part of the communication system, but not reflected light for the return comm.

DJ Kaplan April 21, 2016 at 15:01

I have a question. At this point I am concerned that it will be swallowed up in the mix and not answered, but I will give it shot.
On the Breakthrough Initiatives site, it says that the StarChip will have onboard, among other things, “Photon Thrusters”.
StarChip: Moore’s law has allowed a dramatic decrease in the size of microelectronic components. This creates the possibility of a gram-scale wafer, carrying cameras, photon thrusters, power supply, navigation and communication equipment, and constituting a fully functional space probe.
But the originating study from Y.K. Bae, uses the term “photon thrusters” to mean the source of laser light that is not originating on the StarChip itself:
Photon thrusters amplify thrust tens of thousands of times through a proprietary intracavity system for bouncing photons off of mirrors between satellites.

Please advise.

Paul Gilster April 21, 2016 at 16:24

DJ, I am looking for more information on this. I heard a lot of questions about the nature of these photon thrusters, so you’re not alone in wondering about this. As soon as I get something more detailed on it, I’ll pass it along. The thrusters were not discussed at Breakthrough Discuss unless I missed something.

Alex Tolley April 21, 2016 at 16:38

@ DJ Kaplan, Paul Gilster

From Lubin’s “Roadmap…” paper:

Combined with small photon thrusters (embedded LEDs/lasers for nN thrust steering on the wafer gives a functional spacecraft.

DJ Kaplan April 21, 2016 at 16:51


Comments on this entry are closed.

{ 1 trackback }