My Visit to the SNO

Tom Davis
Last updated: May 19, 2005

My adventure started about one hundred million years ago when the planet earth near Sudbury, Canada, was struck by a meteor in what may have been the largest such event ever. Even today, a hundred million years later, the town and the entire region lies in a "Sudbury depression" that is many, many kilometers in diameter.

Richard woke me up at about 6 am, and as soon as I got dressed, we piled into his car, headed first to a drive-through coffee place to obtain sufficent "nectar of life" to make it through the morning, and then picked up a couple of his friends, Mike and Mark. Richard and his two friends, all from Sudbury, were all very, very excited. And I certainly was, too!

We drove through town and out the other side, and finally stopped at the Creighton Mine, a large-scale, fully-operating nickel mine. Many meteors have a very high nickel content, so that hundred million year old meteor impact at the site of the current Sudbury may simply have buried billions of tons of nickel there and today's mines are simply shafts into what was the interior of that meteor. I was told that the deeper they go, the larger the ore deposits are which would be consistent with going deeper and deeper into the spherical meteor body.

After surprisingly few false starts, we found Fraser, our guide for the day, and signed a bunch of paperwork and legal releases.

We then gathered up our equipment that Fraser loaned us, went back to his office and changed into it: coveralls, iron-toed rubber boots, a heavy belt, safety glasses, a hard-hat, ear sound-protectors, and a large rechargeable battery and light that hooked to the hard-hat. We also double-bagged our cameras in plastic bags.

Then we walked to the mine entrance and waited at the elevator, called "the cage" with about 40 or 50 hard-rock miners. The elevator is a two-story thing, with each part capable of holding 42 men for a total of 84. The doors clanked shut and we started our descent down the deepest straight-line mine-shaft in the world: it goes all the way down in one straight shot to 7000 feet.

To get to our destination, we only needed to go down to 6800 feet, and the whole trip down took about 5 minutes, with three or four stops on the way to let off the miners at various other levels. I think my ears popped about 18 times on the way down, so their advice not to attempt the trip with plugged sinuses was probably pretty good.

When we got to the bottom, the five of us walked along a horizontal mine tunnel that was perhaps 20 feet high and about as wide and was shored up with wire screening where the screen's "wires" were 1/4 inch in diameter with four-inch holes and the screen was nailed to the rock wall with metal rods that went four to eight feet into the rock.

We walked for about a mile, following train-tracks and stepping out of the way of the odd train carrying equipment, and from time to time had to put on the ear protectors because of the howling of the fans that were pumping air into the mine for ventilation. Perhaps 100 yards from the end, the mining activity stopped and the "wire screen" turned to special screening that was made with stainless-steel "wire", so that it would last for 30 years, even in the slightly corrosive mine environment.

At the end of the tunnel, the mud and dirt flooring ended and we were standing on a concrete floor, and there were a couple of high-pressure water hoses to clean most of the dirt and dust off the boots. Having done that, we stepped through a door into an dressing area -- well -- not exactly a dressing area, but an undressing area. At this point we also removed the first bag from around the cameras, discarded it, and passed the bagged cameras through a slot to the other side.

We all stripped naked, stepped into showers, and thoroughly scrubbed ourselves. We also scrubbed our watches our glasses (if we had them) and our protective safety goggles. Abandoning the old clothes and hard hats, after showering, we stepped into another room with the bagged cameras and special "clean-room" overalls, socks, boots, and new hard hats. We were approaching what is perhaps the cleanest area on earth: cleaner than the clean rooms where companies like Intel manufacture silicon wafers.

We had some coffee and muffins in a sort of cafeteria, and then passed through an "air shower": a room that held two of us at a time, and had high-pressure air jets pointing at us from floor to ceiling. From that point on, we must have passed through three more sets of doors, with sticky material on the floor to help collect any dust that might be sticking to the bottom of the boots.

We were finally inside the Sudbury Neutrino Observatory's (SNO's) main lab.

The whole lab is organized around a neutrino detector which is simple in concept and extremely difficult in practice. Every second, our sun emits millions and millions of neutrinos in every direction and the goal of the lab was to try to detect and study a few of them. There are a bunch of good theoretical reasons to do so that I won't go into here, but among other things, it'll tell us whether certain theories about nuclear fusion are correct, and will yield some important information about the structure of the entire universe.

Unfortunately, neutrinos are very difficult to detect. The vast majority of them pass through the earth completely unaffected: even passing through the center of the earth via 8000 miles of rock hardly ever changes their path at all. They have no electrical charge (hence the "neutr" in "neutrino"), and no rest mass, so they travel at the speed of light, and they interact with almost nothing.

Luckily, it's "almost" nothing. On rare occasions, they do interact with deuterium, "heavy hydrogen" that has a nucleus consisting of a proton and a neutron instead of just a proton, so if you watch a bunch of deuterium, you'll "see" a reaction from time to time. Deuterium behaves chemically like hydrogen, so the easiest way to handle "a bunch of deuterium" is reacted with oxygen to make what's known as heavy water, and heavy water is available, since relatively large quantities of it are produced for use in nuclear reactors.

There are a bunch of problems with the heavy water, however. One is the cost: it's about $300 per liter. The SNO detector requires about $100,000,000 worth; you can do the math. Needless to say, in every area where the heavy water was handled, there were giant catchment basins to recapture any accidental spills. Fraser told us that they'd been lucky so far and that the largest spills had been on the order of 10 milliliters or so. The amount of heavy water in the SNO tank amounts to about twice that used in a single traditional nuclear reactor.

The neutrino detector consists of a ball of heavy water surrounded by a shell of regular water in which the detectors are arranged so as to cover almost the entire surface of the sphere. The way this is done is to suspend a large (22 meter diameter) geodesic sphere with 9600 individual detectors arranged around it in a slightly larger excavation deep down in the Creighton nickel mine. Inside the detector sphere is an acrylic sphere filled with the heavy water. The acrylic is about 2 inches thick. There's another containing sphere (I'm not sure what it's made of) outside the detectors.

The detectors (which are about a foot in diameter) are thus mounted between the inner and outer spheres and are immersed in regular water. All the 9600 cables from the individual detectors snake out of the water and come up to a platform over the suspended sphere where they're hooked into an amazing array of electronics: each circuit can handle eight detectors, there are four cards per motherboard, and about 16 motherboards per rack, making about 20 huge racks of electronics.

It was impressive to stand on the platform with all the electronics, not only because of the number of cables, but from there we could view the totally massive supports that were necessary to suspend the giant sphere of water and detectors from it. There are four massive concrete supports around the outside that are anchored to the surrounding rock with cables that go back into the rock to a distance of 80 feet.

Much of the weight is supported by "ropes" of a material I'd never heard of (and whose name I've forgotten) that was chosen since it's stronger than kevlar, and kevlar wasn't up to the challenge.

When a neutrino hits and interacts with a deuteron, a very high-energy photon is released traveling in the same direction as the neutrino had been, which, in the vast majority of cases, is directly away from the sun, wherever the sun happens to be relative to the SNO. Since the 8000 mile diameter ball of rock that makes up the earth stops almost none of the neutrinos, the detection can go on 24 hours per day, but the apparent direction of most of the neutrinos will move around as the earth spins.

When a neutrino traveling at the speed of light in a vacuum strikes a deuterium atom's nucleus, it will produce a photon traveling at the same speed and direction. But the speed of light in heavy water is slower than the speed of light in a vacuum, so what happens is something very reminiscent of a jet plane breaking the sound barrier. The jet is making a sound, but moving faster than sound, so what it generates is a shock wave (usually called a sonic boom) that looks like a cone with the jet plane at the tip of the cone and the taper of the cone dependent on the speed of the jet.

In the same way, the photon that's effectively moving faster than light in the heavy-water medium generates something like a conical "shock wave" of photons that's called "Cherenkov radiation". This Cherenkov radiation appears as a circle of photons hitting the detectors and the size of the circle indicates where in the heavy water bath the reaction took place and in what direction the neutrino was traveling.

The detectors thus need to detect individual photons, and it's done almost like in a standard photo-multiplier: the photon slams into a metal plate and knocks loose some electrons. But there's an electrical field behind that plate that accelerates those electrons to slam into another plate behind it. And the process repeats for a number of stages (maybe 8 at SNO?) with the result that a single photon knocks loose about 20 million electrons in something like a 10 nanosecond pulse that's easily detectable.

Obviously, any photon in the system may cause the odd detector to fire, so there is some elaborate software to determine when "enough" have fired to warrant collecting and saving the data. It turns out that when more than 15 fire in the same 100 nanosecond interval, the data from all the detectors is saved for 450 nanoseconds (a little less than a half of a microsecond) and analyzed. A lot of the collections are trash, but you don't miss too many this way.

In fact, there are only a few real neutrino detections every day, perhaps eight or ten of them. These numbers correlate very well with the theory, although there's still a lot of other work to do. Apparently the inital results only showed about one-third of the expected number of neutrinos, but it was realized that there are three different kinds of neutrinos, and when the experiment was modified to count all of them, the results suddenly matched the theory very well.

In addition to the neutrino detectors, there's also an elaborate set of "neutral-current detector" rods mounted inside the tank, and I confess that I did not understand exactly what they were for and how they worked. I believe it's a sort of cross-check on the neutrino detector data, but could be completely mistaken.

So the idea is pretty simple, but the details are nasty. The big problem is high-energy cosmic rays that are generated in super-novae all over the universe and consequently arrive at the earth from every direction in a more or less uniform manner, and when one of those piles into the detector, thousands of photons are released. If we think of a neutrino-caused event as a flicker of a candle flame, a cosmic-ray event is more like turning on a 10,000 watt flood-lamp.

Luckily, cosmic rays are far more likely than neutrinos to interact with normal matter and 6800 feet of rock does a pretty good job of stopping the vast majority of them. There are still many times as many cosmic ray events detected as neutrino events (maybe 100 times as many?) and of course most of those are coming straight down. As the angle varies from straight down (the shortest path through the earth to the detector), the amount of rock through which they need to pass increases rapidly, and Fraser said that at angles steeper than about 40 degrees, essentially none of them make it, due to the thickness of the earth.

At the surface of the earth, the cosmic ray events would completely swamp the neutrino-caused events.

The other big problem is that there are plenty of other sources of radiation: uranium and thorium appear in normal dirt and dust in levels measured in parts per billion. But there are way more than billions of atoms in any speck of dust, so there's a lot of these elements, and there's a certain percentage of those atoms that have the form of isotopes that are radioactive.

The shell of regular water around the deuterium blocks most of that radiation, but it's better to block most of a tiny number of their decay events than the same percentage of a large number of decay events which explains why the clean-room environment is required.

I guess a worse problem is that the radioactive gas radon is reasonably soluble in water and having such reactions occurring in the heavy water itself would be really bad news, so truly heroic measures have to be taken to remove it from both the heavy and light water. There are incredibly elaborate water filtration systems to do this.

In addition, you really don't want any stray light in the system: a single stray photon can cause a false event, so light-shielding is extensive. But to check the water flow, black tubes would be a nightmare, so they're all made of translucent polypropylene. So those tubes, anywhere near the detector, are wrapped in black plastic.

Due to the law of entropy, extremely pure substances don't like to be that way. Hence, even incredibly pure water "wants" to have some impurities in it, and it is surprisingly corrosive. Many of the normal materials that you normally think of as inert in water (like stainless steel) will dissolve relatively rapidly in the ultra-pure SNO water. Well, rapidly if you're worried about such a tiny level of allowed impurities.

And of course it doesn't do any good to have all these cleaning systems if there's no way to measure just how clean the products are, so there are effectively special chemistry and physics labs just to measure impurities in all the supplies. And standard polypropylene is too reactive with water, so there's a special type of SNO polypropylene, and the rocks at 6800 feet under the ground average about 110 degrees Fahrenheit, which is way too warm, and the list of problems goes on, and on, and on ...

And the totally amazing thing, of course, is that the whole lab is inside a nickel mine: perhaps not the dirtiest environment in the world, but the filth coefficient ranks right up there.

Of course I'd never been in a huge operating mine before, and that would have been amazing enough in itself, but that's another story.

OK, I'll tell one story: how the mine itself is cooled. During the winter, the icy air from the surface is pumped directly down a ventilation shaft, but a fine mist of water is sprayed into it as it goes down. The cold air causes the water to freeze on the walls of the shaft, forming a thicker and thicker coating of ice to form. Then, as the weather warms up in the summer, the warm air from the surface is pumped down the same ice-coated shaft, where it it chilled on the way down. Apparently this scheme almost eliminates the need for artificial air conditioning.

Then there's the question of how to get tens of thousands of tons of nickel ore to the surface each day from a depth of 7000 feet ...

Want to send me mail? Click here:

Return to my home page.