It’s No Small Matter

How do you catch a sterile neutrino?

It’s one of the hardest experiments in physics: an attempt to pin down a particle that leaves no tracks and interacts with nothing -- a ghost as hard to grasp as the chill that raises the hairs on your neck.

The only hint that this subatomic poltergeist even exists popped up in a 1995 experiment brushed off as almost certainly wrong by the vast majority of physicists. If confirmed, however, the finding would shake physics to its boots, introduce a whole new family of particles and perhaps help explain why the universe is made of matter.

So naturally, when Columbia University physicist Janet Conrad invited me to help search for the particle at the world’s premier physics lab, I could hardly pass up the chance. The fact that I can hardly hang a picture didn’t faze Conrad a bit. It’s just like cooking, she said. “There’s a recipe by which you put it together. It comes out, or it doesn’t.”

Except that the kitchen in this case was Fermilab -- a 6,800-acre complex of giant particle accelerators 45 miles west of Chicago.

As someone who’s spent dozens of years peering at physics from the outside, here was an opportunity to get down and dirty (literally, as it turned out) -- an embedded journalist on the front lines of physics.

Physics in the flesh is nothing so neat as the crystalline world of equations, nothing so simple as the myth of Newton getting bonked by an apple and -- bingo! -- beholding the secrets of gravity.

It is a world where everything’s a mess, where you can’t see what you’re looking for and what you can see is harder to pick out than a soft breeze in a storm of uncertainty.

Even ordinary neutrinos are so insubstantial they can slip through a light-year’s worth of lead without jostling a single atom; a changing cast of trillions occupies your body every second. Yet they are weighty enough that they won two physicists the 2002 Nobel Prize: To everyone’s amazement, it turned out that two-thirds of the neutrinos expected to make their way to Earth from the sun appeared to be missing, and the reason they were not detected seems to be that they are morphing into different forms en route.

According to prevailing models of physics, neutrinos have no mass. But this shape-shifting means they must have a tiny bit of heft -- because in order to change identity, they also have to change mass.

This smidgen of mass matters because neutrinos outnumber all other particles a billion to one, pouring out en masse from the nuclear furnaces of stars, radioactive atoms in the earth and cosmic ray collisions in the atmosphere; they outweigh a universe of stars.

Sterile neutrinos, if they exist, are expected to weigh even more, enough to divert the streams of galaxies that flow across the sky.

Alas, it’s called “sterile” for a reason. It’s so innately unsociable that it couldn’t communicate its presence if it wanted to.

Tracking down such a pathologically shy particle requires an exquisitely subtle (even sneaky) experiment; lucky for me, it’s going on at the coolest particle physics laboratory in the world.

No matter how often I visit Fermilab -- officially Fermi National Accelerator Laboratory -- it never ceases to amaze me. Seeing the Fermilab high-rise suddenly sprout from flat Illinois farmland seems as unlikely as coming across a cathedral in a cornfield. In fact, Fermilab’s founder, the late physicist and artist Robert Wilson, talked of such accelerators as the “cathedrals of contemporary science.”

For particle physicists, Fermilab is the Vatican.

To enter the site, you pass under an asymmetrical three-legged arch -- one of Wilson’s many sculptures. Farmers still lease the land, growing corn and beans. Huge machines designed to re-create the Big Bang burrow almost invisibly under wildflowers and prairie grass -- the playground for hundreds of species of birds, beaver, weasel and mink.

As a toast to the country’s literal frontier, Wilson installed a herd of buffalo. They usually stand near their barn looking bored.

The long road into the heart of the lab traverses a serene landscape of fields and ponds, ending at a reflecting pool presided over by a phalanx of flags from dozens of countries. Behind looms the high-rise, which Wilson modeled on Beauvais Cathedral in France. Instead of stained glass, clear windowed walls swoop 16 stories high. A multilevel atrium, lush with trees and hanging vines, encircles a giant brass pendulum that seems to fall from the sky.

I’ve been coming here for nearly 20 years but always as an outsider. This time, I am granted a badge of belonging that lets me wander into the inner sanctum -- the main control room, the soul of the machine. Inside, technicians choreograph the paths of protons as they loop their way around a complex of accelerators with a precision that makes circus plate twirlers look like amateurs.

By the time they reach the four-mile round collider called the Tevatron, the protons are traveling at nearly the speed of light. And when they crash, the concentrated energy sets off fireworks of exotic particles -- mini Big Bangs.

“I’m always amazed that the protons even get around the ring,” Bonnie Fleming, one of the physicists in our group, told me. “It’s a Rube Goldberg machine of astonishing complexity.”

‘A Gnat’s Whisper’

Anything can jam the works. One week, an air conditioner turning on caused a power surge that threw off a magnet. The week before that, surges caused by lightning shut down the accelerator for the entire weekend.

Our search for the sterile neutrino involves an even more delicate dance of particles.

Neutrinos are so elusive that the first physicist to detect them, in 1956, said it was like “listening to a gnat’s whisper in a hurricane.”

Over the next 40 years, a total of three kinds of neutrinos were discovered: the electron neutrino, the muon neutrino and, three years ago, the tau neutrino -- each paired with a member of the electron family (muons and taus are heavier versions of electrons).

But an experiment at Los Alamos National Laboratory in 1995 hinted that there might be a fourth member of the family -- the sterile neutrino. In effect, the experiment found that there were three differences between masses of neutrinos, which would require a total of four kinds of particles -- just as there are four spaces between the five fingers of your hand. (To have a fifth space, you’d need a sixth finger.)

Alas, almost no one believed the experiment was right. The few exceptions included physicist Bill Louis of Los Alamos and eventually Conrad -- the two co-leaders of the current experiment. When they set about to confirm the Los Alamos findings, colleagues told them they were ruining their careers.

Neither Conrad, with her girlish Midwestern looks, nor the slim and soft-spoken Louis, appears much like a troublemaker. But should the Los Alamos experiment be confirmed, “then it turns the whole world of physics on its head,” said Fermilab physicist Joe Lykken. “Everyone is betting their homes and first-born children that it’s wrong.”

Most neutrino experiments passively wait for the particles to rain down (or up) from cosmic sources. But the Los Alamos experiment created a controlled beam of neutrinos to track what happens over a precise distance. So that’s what Conrad’s experiment must do.

Our first order of business, then, is to create a well-calibrated beam of neutrinos. If you want to know what comes out of an experiment, you have to understand what goes in.

Like all particles produced at Fermilab, the process begins in a small bottle of hydrogen buried inside what Conrad calls a “Frankenstein machine.” Long metal legs braceleted with fat silver doughnuts stand several stories tall -- R2D2 on steroids. I expect to see lightning bolts fly out -- and that’s not so far from the truth. The beast electrifies the hydrogen, then sends these “ions” off with a 750,000-volt push.

From there, a magnet bends them into the LINAC, or linear accelerator. Inside its shiny copper cavities, the ions surf on radio waves, getting pushed to ever higher speeds.

The LINAC hands the beam to the Booster, which strips off electrons, leaving naked protons. The small circular accelerator spins the protons around 50,000 times, adding speed with every turn.

The noise in the Booster tunnel is painfully loud: There’s the thump of vacuum pumps, the banging of metal as magnets turn on and a loud, woodpecker-like “rat-a-tat-tat.” Each “tat” is a “spill”; each spill has 84 bunches; each bunch has 60 billion protons.

Our neutrino experiment gets five spills per second, delivered in short pulses, straight from the Booster. Thus, its name: MiniBooNE, for “mini” Booster Neutrino Experiment. With some 60 scientists from 12 institutions, it’s the smallest experiment in high energy physics.

Most of the protons from the Booster go on to the Tevatron. Ours peel off and crash into a beryllium target, creating unstable particles that drift through a 150-foot sewer pipe where they disintegrate into a mix of neutrinos and other particles. A highly magnetized horn focuses the beam, rattling in tune to the same loud rat-a-tat-tat; the sound follows the whole experiment like a heart beat.

At the end of the line, a steel wall absorbs everything that’s left except neutrinos.

Conrad takes me across a muddy field, tracking the path of the beam underground, until we come upon what she describes as a “home for Teletubbies.”

It is large mound of dirt with prairie grasses sprouting on top like a bad haircut. A door leads to a room full of computers and power sources -- all marked with signs reading: “Danger! High Voltage!” The muffled rat-a-tat-tat of particle bunches follows us like the trill of some strange bird.

Under our feet is a sphere filled with 800 tons of baby oil so clear you could see a 100-watt bulb from half a mile away. If we’re lucky, it’s here that the sterile neutrino will give itself away.

The Calling Card

Of course, you can’t see a neutrino. But on rare occasions, neutrinos collide with other particles, leaving calling cards you can see. Electron neutrinos produce electrons; muon neutrinos produce muons.

When an incoming muon neutrino from our beam collides with a carbon atom in the oil, it will produce a muon most of the time. But if the Los Alamos experiment is right, a small fraction of the muon neutrinos will change into electron neutrinos. These neutrinos aren’t detected directly. But the electrons they produce show up as rings of light that are “seen” by 1,520 amber phototubes that line the tank like harvest moons.

If enough electrons show up in the data, then the Los Alamos experiment is confirmed. The reasons are mind-numbingly complex, having to do with the relationship between neutrino mass and distance traveled. But the bottom line is that if the electrons come where MiniBooNE is looking for them, then there really are three gaps between neutrino masses, meaning there really is a fourth neutrino -- the sterile.

I’m still high on all this power and grandeur as Conrad drives me to my work site. We pass a half dozen Wilson sculptures, including an orange and blue “capacitor tree,” a building that looks like a pagoda, and another whose staircase is modeled on a double strand of DNA. The road runs by a long string of power lines, which Wilson had constructed in the shape of the symbol for pi.

It’s disillusioning to say the least when I’m taken into a cement structure that looks vaguely like a prison.

Inside, catwalks circumnavigate a huge bare hangar several stories high. The floor is covered with abandoned electronics racks. Everything is orange and blue, Fermilab’s colors.

In a small, dingy room off to the side, I meet my mentor for the week, Len Bugel, a physics teacher from Stratton Mountain School in Vermont who has been helping out here over summers for nearly 10 years. His car has an “I love neutrinos” bumper sticker.

I’m also working with a Columbia University physics undergrad, Clarisse Kim, who confesses that she hated science until her mind was changed by popular books and articles -- some, I’m delighted to discover -- mine. Now, she’s teaching me.

My first assignment doesn’t seem very promising, either. Bugel brings out some ratty old particle counters -- clunky aluminum constructions covered with torn black paper and peeling tape. “I retrieved them from the dump,” Bugel says.

In fact, nearly everything in MiniBooNE has been used somewhere else before. “We built it out of junk,” Conrad says. “But all the junk works.”

To ensure that our recycled counter doesn’t pick up any stray light, we tear off the old black paper and tape, cut new paper and tape it on as well as we can. The counter is a jumble of mismatched shapes, and I wind up using almost a whole roll of tape just trying to secure the seams. The rubbery black tape is hard to cut -- and very sticky. Tar-like goo gets all over the scissors, my hands, my hair.

Life in ‘Prison’

Even this kindergarten-level task has me feeling like a klutz.

As Conrad leaves, she hands me some papers -- my “cheat sheet,” she calls them. Actually, they’re four single-spaced pages of marching orders. The work includes “light-tighting” and “plateauing” counters, designing “telescopes” and a bunch of other stuff that doesn’t make any sense.

Late that afternoon, I sit through a talk on a subject I thought I understood well -- how cosmic rays from space produce showers of particles that turn up as false signals in our experiment. But even though the lecture is for undergraduates, most goes over my head.

Conrad tries to console me by pointing out that the other students have already had a dozen lectures, and so can speak the language of “hodoscopes” and “paw-plots,” “Monte Carlos” and “Michel packages.”

It’s a good thing I have Bugel as a guide. Back at our “prison” the next day, he shows me how to test whether our particle counters are working correctly.

This scut work is more important than it seems: MiniBooNE has to account for every muon and every electron produced inside the tank. But anything that comes from outside the tank is potentially confusing “background,” or “yuck,” as Conrad calls it. A separate shell of phototubes lines the tank, creating a “veto” region that keeps track of muons leaking in from the outside. That way, they can be subtracted from the data.

But how can we know that the veto counts all the interlopers?

The only way to be sure is to lay down extra counters below the tank. Inside of each of our junkyard counters is a large square of plastic infused with fluorescent material that gives off light when hit by a muon. The phototube turns the light into electrical signals.

Bugel mixes five-minute epoxy to glue on the phototube, and we untangle a knot of red and green wires (red for high voltage going in; green for signals coming out). We get a sharp peak on the oscilloscope, suggesting that muons are raining through the ceiling.

Success! I may flunk undergraduate physics, but I’m OK at paper and tape. When it’s time to test the next counter, I get to mix the epoxy.

Still, the work is terminally frustrating.

We glue on a phototube only to realize that a metal sleeve that protects the seam from stray light doesn’t fit. We have to take off the tube, get off the glue. But someone has taken the alcohol. Someone also has taken the paper towels.

We manage to peel off the glue and try again. This works, but this time our signal isn’t nice and sharp. After ruling out problems with cables, connectors and power supply, Kim decides the phototube must be bad.

We untape everything, but now the glue has set. Kim patiently chips away with a single edged razor, and predictably winds up with a nasty cut. We have first blood.

We find another sleeve and start again from scratch.

Each counter requires the same drill. And then we need to test them again -- this time stacked in pairs. To get accurate readings, we need cords of the same length, so Bugel scrounges for two 16-nanosecond cables. (Length is measured in the number of seconds it takes light to travel through.)

Channel One (from one counter) is counting fine, but Channel Two (from the other) isn’t. Now what?

We take everything apart and start all over again.

I complain to Fleming about how slow things are going. She’s surprisingly sympathetic, given that she herself had to calibrate all 1,520 MiniBooNE phototubes. “In a normal job, you might say, ‘I worked all day and didn’t accomplish anything.’ In this job, you say, ‘I worked all day and took a step backwards.’ ”

Troubleshooting, she says, is what takes up most of your time. “And it’s not like you can call somebody in to fix it. You’re the person who has to fix it.”

This is nothing like my kitchen, where an occasional oops might topple a souffle. Here, whether or not we find the sterile neutrino will turn on these tiny details -- tape and phototubes and glue.

Pitfalls Abound

You can’t afford to miss a thing when you’re looking for a needle in a haystack, and this is a whole lot harder. “You can get rid of 98% of the hay, but everything you’re left with looks just like needles,” Conrad says when I join her a few days later for her Sunday 8-to-4 shift.

For every “real” electron signal, 100,000 are produced from cosmic rays, and the only way to sort them out is to match the timing of the outgoing signal with the incoming pulses of particles -- the rat-a-tat-tat.

All this activity is monitored in the computer-packed MiniBooNE control room on the 10th floor of the high-rise. Every two hours, Conrad checks all systems, entering everything into an electronic log. A little red heart means all is well. A line across the heart means trouble.

As we trek back out to the Teletubby mound for the once-a-shift site check, I’m surprised to learn that at least some potential troublemakers are familiar. “You never know what animals are going to get in there,” Conrad says. “The most important thing you do is listen and smell.”

She checks the oil level in the tank, looks for leaks, studies the banks of computers. She checks the amount of current going to each phototube, the readings for temperature and humidity for the last 24 hours. She checks the lasers that flash six times a second to make sure the phototubes are working properly.

“It’s so easy to make a mistake,” says Louis, listing just a few of the ways: There could be a bug in the software, a glitch in the electronics or background noise that you forgot about. “You want to look under every rock and see if there’s a scorpion under it,” he says.

Even if the experiment performs perfectly, there will always be some uncertainty about what is needles, what is hay. The best you can do is learn to recognize the hay as best you can and gather enough data for solid statistics. It’s somewhat like flipping a coin. If you get heads 10 times out of 10 throws, it could easily be a fluke. But if you flip a million coins and get a million heads, you’ve stumbled on something truly weird.

The need for good statistics explains why MiniBooNE’s results won’t be known for several years. “Maybe it’s a little like falling in love,” says Conrad, who is married to a physicist. “It’s slow steps forward to the point where you’re actually confident.”

MiniBooNE is under a special onus because the implications of a sterile neutrino are so big, and also because so many people are so skeptical. “The more significant the result, the more proofs required,” Louis says.

It’s taken almost two weeks to test and calibrate Kim’s counters. But finally, our big day comes. Bugel has gotten all the counters into the MiniBooNE detector pit. So along with Louis, we go down to take a look. And since the pit is an ODH (oxygen deficiency hazard), we first have to sign a book and call the Fermilab fire department to let them know we’re going down; one person has to stay above -- just in case. (In training, they tell us the most dangerous thing you can do is go down to rescue a colleague.)

Hazards lurk everywhere.

A sign hangs on the chain blocking the metal hatchway: “Danger! Confined Space. Keep Out.” Kim pulls up a rickety ladder, and we descend into the belly of the beast on grilled metal stairs. The staircase curls around the stark white tank, which looks like an enormous soccer ball suspended on fat legs. It’s dark and spooky. There are more “Danger” signs below.

Playing with big machines is nothing like sitting at a desk diddling with equations. Compressed gas cylinders can “become torpedoes,” they warn in safety training; magnets can make projectiles of tools.

Our counters are now lying under the tank, and we’re supposed to connect them to the power sources and readouts in the room above.

But our 16-nanosecond cables won’t reach. We get longer cables only to find that they won’t fit through the hole in the floor. The hole gets enlarged, but longer cables mean longer light travel time, which changes everything.

I now understand why you see clots of construction workers on the street, seemingly doing nothing: I imagine them waiting for cables to be cut, holes enlarged, equipment recalibrated.

Suddenly, Kim’s oxygen monitor starts beeping, flashing red lights. Are we all going to pass out? We scramble upstairs -- only to find out she had her finger over the air intake valve. (We later discover that the monitor also goes off in the high-rise elevators.)

So much worry and work to tease out a probably nonexistent ghost. It’s hard not to wonder why physicists bother.

One reason, says Fermilab’s John Beacom, is that there will be no progress in neutrino physics until the results of the Los Alamos experiment are clarified. “[It’s] going to be clanking down the halls of neutrino physics forever until it’s put to rest.”

Another reason is that a fourth neutrino could force physicists to rethink everything from the mechanics of the Big Bang to the creation of elements. It is so odd that it may even have a different origin than the other neutrinos. Perhaps, says Lykken, it somehow arises from extra, hidden, dimensions.

“The discovery of sterile neutrinos would be an even bigger surprise than the discovery that neutrinos have mass,” he said.

And the fact that neutrinos are overall oddballs (it is the only particle that twists exclusively to the left, for example) leads physicists to think they might be the key to other fundamental mysteries: Is the fact that the universe is made of matter when antimatter should be present in equal amounts due to a similar cosmic twist?

Even if there’s no signal whatsoever -- say, if the Los Alamos experiment was wrong -- MiniBooNE is still considered a crucial experiment, because understanding what doesn’t exist is every bit as important as understanding what does.

“It’s like chasing a quarry into a corner,” Conrad says. “If we kill [the Los Alamos experiment], we’ll be heroes. And if we find sterile neutrinos, then we’re really heroes. Either way, we’ve made a big step forward.”

I’ve got just two days left, and one more pair of counters to build. Bugel hands me a plastic bucket and says: “Let’s go shopping!”

Down on the concrete floor of the huge central cavern in our orange and blue lab, we ransack boxes for screws, nuts, connectors. Bugel has already cut green heavy-metal support struts to the lengths we’ll need.

Putting them together makes short work of fingernails. Experiments depend as much on sweat and blood as magnets and cable. I learn a lesson that’s apparently well-known in body shops (one likely to come in handy at home): “Bang to fit; paint to hide.” We do a lot of banging to fit, but since this construction will sit underground, we don’t have to paint.

By the next day, Conrad has sent me a photo of our clunky green construction sitting in place underground beside the proton beam. I feel like a proud mama. I’ve added one small piece to a mosaic that might help solve a cosmic-scale puzzle.

At the end of the day, even the most elaborate experiment is made up of a lot of little nothings that add up to something grand.

A lot like the neutrino itself.

Before saying goodbyes, I rollerblade one last time around the Tevatron along a river of cooling water graced by Wilson’s fanciful fountains. The warm water attracts a blue heron.

Paying Respects

My last stop is the tiny pioneer cemetery where Robert Wilson was buried in 2000. The other gravestones all date from the 1800s.

Standing by Wilson’s plot, I can almost reach out and touch the high summer corn. Not far away, reddish baby buffalo stumble about on still-wobbly legs. In the distance, Wilson’s cathedral stands alone, beseeching the sky for answers.

And deep underground, the sterile neutrino is (perhaps) preparing to stage an appearance.