With every thing Mark Suppes fixes, another thing breaks. We're in his workshop, and he's hunched over, tightening the bolts on a squat tubular machine. This is the fusion reactor Suppes built, and it's not working.
Suppes never slows down, moving from one problem to the next with an irrepressible smile. The workshop is a few hundred square feet sub-let from a roboticist friend in a warehouse one floor above a hassidic clothing factory near Bed-Stuy in Brooklyn. "I'm starting from nothing, I mean nothing," says Suppes, "There's no reason I should be doing this. It's ridiculous on all levels." What he's doing is building a Bussard Polywell fusion reactor.
Dr Robert W. Bussard was a a physicist at Los Alamos, a rocket builder in the age of rockets. His proposed hydrogen-compressing ramjet became the stuff of sci-fi legend. (There's one of the front of the USS Enterprise.) He was a classic scientist of the 20th century retro-future, when it was all going to be clean atomic energy, spaceflight, and freckled white people living on the moon. It didn't happen that way, and in the end he died an old man at the bottom of the same polluted and overheating gravity well he was born on.
But in the few weeks before his last lab shuttered from lack of funding, he'd had a breakthrough, and Bussard believed that he might have solved the most difficult physics problems of fusion energy.
Fusion reactors work by creating very high energy plasma inside of a vacuum, and then getting small atoms close enough for the attractive force of their nuclei to overcome the repelling force of the atoms' natural electrostatic repulsion, causing the two atoms to fuse into a new element. A lot of energy is released when this happens.
The problem with fusion has always been that we don't know how to get more energy out of it than we put into it. We know the energy is there. We know effective fusion is likely to take a lot of energy to jumpstart, but we don't know how (or if) we can ever get fusion going well enough to capture as much energy out as we put into it - the elusive break even point. We can't control how the energy likes to leave the reaction, and it's in forms we can't use. We can't keep the reaction self-sustaining like it is in the sun, since we don't have gravity on our side. Also, we're talking about plasma, the star-hot fourth state of matter. It tends to destroy the equipment.
Dr Bussard was one of the many government scientists trying to make a workable fusion reactor. He'd been one of the forces behind the tokamak, the main ongoing avenue of fusion research in the US. But Bussard eventually wrote a letter blasting the US fusion program as expensive, impractical, and doomed in the 1990s, testifying to Congress against much of what he'd once advocated.
In the last decades of his life Bussard left the government and founded a fusion company called EMC2. "He tended to jump from project to project. Today he might be diagnosed as ADD," said Tom Ligon, who worked with Bussard for years as his "hands in the lab".
Bussard was frustrated by a string of failures and funding problems. EMC2 was mainly supported by the Navy, and by their rules he couldn't publish his research. Eventually EMC2 lost out in a round of budget cuts. Bussard's team worked right to the end, building a 6th version of his "Polywell" reactor design. In the last weeks WB-6 looked promising, but the reaction burned through the insulation and destroyed it. After the lab closed Bussard analysed the last of his data. He came to believe what they saw in the few runs before the WB-6 self-destructed was what he'd looked for all along: a clean fusion reactor, which, if scaled up to about three metres, would change absolutely everything. But he couldn't know for sure until someone built it. "From an outsider's perspective there were still some physics questions to be answered," says Ligon. "I will guarantee you he squeezed everything he could from those neutrons!" (The presence of neutrons is used to detect fusion.)
Bussard began a crusade, telling everyone who would listen about his results, looking for anyone, governments, corporations, schools, that would finish the Polywell. He did a Google Tech Talk, and challenged Google to take it on. "He told me he was starting to feel a little ill the day of the Google video talk," says Ligon. "He discovered he had a second cancer the day after the Google video. He really was desperate to get the thing funded again."
Dr Bussard died not long after, at 79, his life's work yet unfinished. But the Navy did step in and fund EMC2 for two more years with stimulus funds. "Dr. Bussard died knowing the project had been picked up," says Ligon. But the work is back under the Navy embargo.
* * *
By day Mark Suppes is a Ruby on Rails developer for Gucci, and a serial startup guy. He saw Bussard's Google tech talk on Youtube, and decided to take up the project himself. He was undaunted by being one guy, not a physicist or even an electrical engineer. He understood just enough to decide this would be a good way to spend a big chunk of his life. He publishes everything he does as open source and blogs his progress.
"The stakes are just unbelievably high ... you [would be]basically the progenitor of a new era of humankind," says Suppes. "If this works, the price of electricity will drop to 10 cents on the dollar. Maybe even better than that. It will be the end of oil and coal as an energy source."
Bussard created a magnetic grid, or magrid, out of a box constructed of electromagnetic loops. The magnetic field repels the fusion material, but there's a patten in the centre where the fields from the various magnets cancel out. This creates a virtual magnetic container for the reaction to happen in.
Suppes has built his first test magrid out of Teflon and copper, though he hasn't run it yet. He's started designing a 3D printable magrid with space for superconducting magnets, which potentially could take less energy to run and get the reaction closer to self-sustaining. He's using a high temperature superconducting magnetic tape, but even high temperature means liquid nitrogen cooled, instead of liquid helium. It has to sit next to plasma.
"It's the McDonalds problem. How do you keep the hot side hot and the cool side cool?" says Suppes "It's going to have to be a multilevel cooling system... Multiple layers of vacuum mirrored insulation."
"It would be hard to believe you could advance on what Dr. Nebel and Dr. Park (of EMC2) are putting into it," says Ligon. But their funding only goes to next year. Suppes doesn't have institutional support, but he also doesn't have institutional constraint. "I expect to be working on this project for the next ten years, and that's what it will take at least. I have a long term commitment to this," says Suppes, "I would rather really go for something amazing. Even if it doesn't work, I'm learning everything I've always wanted to know about physics, and electrical engineering."
It's a humid 32C Saturday afternoon in Brooklyn when I first meet Suppes at his workshop. It's a chaotic floor full of heavy gear. His equipment is mostly acquired from eBay - decommissioned bits from old semiconductor labs and the like. An 86kg capacitor sits on the floor near a rotor that looks like a small jet engine. Rack-mounted high voltage is hooked to a mobile switch stand with a big red 'stop' button on top. The high voltage unit is working from the a standard 110-volt wall outlet, but stepped up to dangerous levels. Despite being a few feet from a nuclear reactor, the high voltage is the most dangerous thing in the room. The reactor itself is a vacuum chamber mounted on a vacuum pump, with a mass injector connected to the deuterium bottle that pumps the hydrogen isotope into the vacuum chamber.
A data acquisition card lets Suppes control the setup from a PC, and get readings about voltage and vacuum. As for detecting fusion, that's up to a fluorine detector, a large glass capsule of clear gel taped to the top of the reactor's window. If a neutron flies through the gel, a bubble will mark its passing. This can happen naturally from cosmic rays, but it doesn't happen too often. Getting a few bubbles in one night is statistical evidence you've got a fusion reaction going, as the neutrons are thrown out of the fusing hydrogens. Or there's something terrible headed for the Earth, and it all doesn't matter anyway.
The vacuum pump isn't working when I show up, but after a while running, it appears to fix itself. His controller crashes every time the plasma field becomes unstable. After disconnecting it, he realises that it's crashing from EMF through the air. Eventually I put down my camera and journalistic distance and get down to trying to fix the damn machine. We put it in a rigged up Faraday cage, it still crashes. Finally we notice a USB hub that's fritzing out when the plasma field flickers, and take it out of the loop.
Everything is working now, except the software. Suppes attaches the mass injector to a big dial and a 9-volt battery. This works better than his software controller ever has.
As evening falls, Suppes wonders aloud if cheap fusion energy is even a good idea. "Would we just use up the planet quicker?" he asks, then shrugs and moves on. It's a good question. In every age access to easier energy has gone along with environmental destruction. From Native Americans fire clearing forests and the extinction of Australian megafauna to the Industrial Revolution fouling up everything else, we've always used energy for beating the hell out of the planet. Today tens of thousands of barrels of crude oil have poured into the Gulf of Mexico. Might as well try for the devil we don't know.
This Farnsworth fusor is finally working. The plasma is the kind of beauty sci-fi special effects are trying to reach. It's otherworldly, a tiny purple star in a tiny cage.
Suppes asks me to keep the chamber going while he runs out to a late night bodega. I take over pressing the big red button. Turn it on, let it go until it starts to flicker from instability. Shut it down and let the red glowing inner grid cool. Every so often, I check for a bubble in the fluorine neutron detector. It's late enough for Brooklyn to be silent, punctuated by the occasional siren. It's dark. The fusor seems warmed up now, humming and going for longer stretches each time. It dawns on me what I am doing. Press button, stand back, sipping water and watching the purple glow of the chamber. I'm fusing the atom. I'm smashing hydrogen together at unthinkable temperatures and making helium. This is the process behind all life, behind the beauty of the universe. This is the zygote of starkind I am making and shutting down repeatedly.
The ghosts of neutrons
Suppes comes back and asks: "How was it?"
"Weird," I reply. I open my mouth to say more, but nothing comes out. I can't explain it, and he already knows.
The hours roll on, and we start to see bubbles in the fluorine detector - the ghosts of the passage of neutrons. Suppes is drinking Corona from the Bodega. Pretense exhausted, he starts showing me how far he can pop the cap off a Corona. He says it's the best beer for distance. Four bubbles. He's achieved fusion.
Suppes seems to understand his odds of success aren't great, but he's failed before in the startup world, and it doesn't seem to bother him. "I have a very good sense of what it takes to not die," he says. "To me the biggest risk is not taking the chance that could be your life's work. That would really be a shame."
Suppes believes in fusion. "It doesn't need to be me," he says, "I'm just a force of will." But for now, it's him, and one lab in Santa Fe. Next year, it might just be him. He looks back at his fusor. "The longer it goes the more real it becomes."
Illustrations: Wendy MacNaughton