(no subject)
A few people have asked exactly what I'm doing at work here at CERN. I also tend to get the question of 'what exactly does CERN do?' a lot. I'll try to give some answers for those who are interested without, I hope, boring everyone in the process.
Some pictures of my detector
To start out, C.E.R.N. stands for 'European Organization for Nuclear Research' (it is in French) and it is just that, a large European laboratory. The CERN facility is structured something like the Los Alomos National Lab in New Mexico, or in other words, sort of like a college campus. Walking into the lab mostly you see rows of office buildings and then rows of industrial buildings stretching out over the countryside. In actuality there are several separate laboratory sites around the area. My office is in the largest of the sites, and most of the smaller sites just contain a few buildings.
CERN, being a large laboratory, has many experiments being carried out at the moment. Some of these experiments require very large machines, such as the experiment I am working on. The largest of these is the LHC, or Large Hadron Collider, which is a huge particle accelerator. From the surface, the LHC is invisible because it is buried all underground, much like a subway tunnel. This tunnel runs in a large circle that is 27km around and can be hundreds of feet underground. Inside the tunnel is a large pipe that is about 1 meter in diameter and runs around the whole length of the tunnel. By the time the pipe is finished being installed it will be 27km long and the two ends will connect. Inside the pipe is the heart of the accelerator, which I'll get into a bit later. This accelerator is one of several on the CERN campus, but is easily the largest (the next biggest is a 10th of the size or so). Hopefully in about a year, everything will be finished being built and will get turned on. Don't expect to hear from me very much after January 2008 ;)
There is another part to this experiment which is the detector used to tell what happens with the collider. There are 2 big detectors and several smaller ones located at various points around the circumference of the tunnel. The largest of these detectors is called A.T.L.A.S (A Toroidal Lhc ApparatuS) and this is what I work on. ATLAS is big, really really really big. I recently was able to go down underground and watch as it is being constructed and it is just staggeringly... big. It is shaped roughly like a cylinder that is 45m long and 25m in diameter.
The ATLAS detector is made of quite a few diff rent pieces that do different jobs and they are layered like an onion. From the inside out the pieces are the: pixel detector, SCT, TRT, 2 calorimeters, and a muon detector. One of the projects I have is working on some of the software dealing with the pixel detector. Before I can really give any more details on what I do, I should try to explain some of the physics we do.
Physics is really just the study of how the world works. However after many years of study, we know how all of the basic things in the world work more or less. This means that to study new things we have to make use of new technology to study parts of the world that were previously out of reach. For instance computers now can calculate difficult problems quickly enough that we can use them to try to understand VERY complex systems, bigger telescopes let us see the universe in much finer detail, and bigger colliders let us observe much smaller things.
Ok, that last one may take a bit of explaining, here's the analogy that gets used a lot. Imagine that there is a very dark room which you are not allowed to enter, but that there are doors on all sides. You think there might be something in the room, but you aren't sure, how can you tell? Imagine if you took a basketball and threw it into the room a lot of times. If it always sails through, you can guess there is nothing there. If it bounces back at you, then chances are, there is something it is bouncing off of. But can you tell if it is a big square block or a pyramid shaped one? Imagine you throw the ball in hundreds of times and measure where it passes through and where it bounces off, you can start to get an idea of the shape of the object. Now what if you wanted to know if the corners of the object were rounded or not? Your basketball probably couldn't tell you, but what if you used a golf ball and threw it in thousands of times? The idea is that you can only 'see' details which are bigger than what you are throwing in. Luckily for us, we don't usually have to throw balls at things to see them, we can use light. Light, however, works in almost the same way, a light particle (photon) bounces off of an object and back to your eye, billions of them at a time. Light particles are pretty darn small, small enough that it usually doesn't matter to us. However what if we want to look at things that are smaller than a photon (ok, smaller than the wavelength of the photon)? The next step is to use an electron microscope. As the name implies it uses electrons, not photons, which, have a much smaller size. This is great until you want to look at things even smaller, things much smaller than an atom. Well here comes the cool physics trick, the more energy an electron has, the smaller it (it's wavelength) is. This is one of the things quantum mechanics tells us, and if you know of a good analogy for it, let me know, otherwise you'll have to take my word for it. This is one of the reasons we build really big accelerators, to give a particle a lot of energy so we can use it to probe really small things. Really small things = the insides of atoms.
The other reason to build big accelerators is so that we can see new things. One physics equation which most people have heard of is E=mc^2. This says, energy = mass * the speed of light squared. The speed of light is just a number (and a really big one), so this says that energy equals some amount of mass. In other words you could say 10000 units of energy = 1 unit of mass. In a nuclear bomb we change a little mass into a lot of energy. At CERN we do the opposite. Particles (protons here) go through the accelerator until they are moving very fast, that is they have a lot of energy. We can speed them up going clockwise and counterclockwise around the accelerator ring. Then we can let the two protons run into each other (which is why it is sometimes called a collider). Some of the energy the protons has can be changed into mass. What sort of mass? New particles which, in a manner of speaking, come flying out of thin air and pass through our detector! Some of these particles don't live very long at all (a billionth of a second or less) before they decay. Heavy particles tend to decay into lighter particles which decay into lighter particles etc. Eventually we are left with the lightest particles which is what the universe is mostly made up of. These light particles we can study all we want, but if we want to study the heavy ones, we have to make them. Making these particles is the other big job of the LHC, and studying them is the job of ATLAS.
So what do I do here? I have two projects that I am working on. First, as I said, I am working on some of the software for the pixel detector. One of the jobs of the pixel detector is to be able to tell when a particle is traveling through it, and in what direction. To do this, it has three layers of 'pixels' each which can tell if something passes through it. Then a path can be extrapolated by drawing a line through these three points. At the moment the detector is being tested by letting cosmic rays pass through it. These are particles which come from stars and pass through the atmosphere. Usually you can't tell when they pass through the earth, but this is what our detector was designed to do. So every once in a while (once a minute or so), a cosmic ray will pass through the layers and (hopefully) be recorded. I say hopefully because the test is still being set up. The software I am writing has the job of trying to see how well all of this is working. For instance, does it measure particles passing through it, do the lines get connected correctly, is the power turned on? The goal of this is that we well be able to work out a lot of the bugs this way and things will run much more smoothly when we start for real.
The other project I am working on is with a group to try to detect one specific particle. This particle is called the 'single top quark' and is somewhat of an elusive guy. The problem with detecting it is that it is one of those particles that doesn't live very long. 0.0000000000000000000000001 seconds or so (yes I put the right number of zeros in). Because it doesn't live very long, it never makes it into our detector. This means that we can only detect the things it decays into, not the particle itself. The downside of this, is that some other particles happen to look very similar to the things this decays into. For instance, say the 'single top quark' turns into 'b', 'w', and 'v' particles when it decays. How can we tell this apart from the case when a 'b', 'w', and 'v' are created in the collision? Or what about when the 'w' is created in the collision but the 'b' and 'v' come from the decay of a 'h'? There are a lot of these cases, and we have to be able to figure out which ones really matter to us. It's not an easy thing to do. You can measure things like the angles between the 'b', 'w', and 'v', or how fast they travel and try to find differences, but that only gets rid of some of the 'pretend' single top quark signals. My work there, along with several other people is first to try to isolate the true single top events. Another thing to think about, we've never seen one, so how can we be sure we know what we are doing (this keeps grad students like me up at night)? We also work on trying to figure out how much error there is in our methods, and how likely it is that events are single top quark events. Lastly, if we can successfully do the first two, we can measure properties about the single top quark.
Practically speaking, I spend my days on the computer writing software, debugging software, reading physics papers, running my software trying to figure out how to isolate single top quarks, re-writing my software to do a better job of it, eating lunch at the cafeteria, and drinking too much espresso (which is really really good here).
Some pictures of my detector
To start out, C.E.R.N. stands for 'European Organization for Nuclear Research' (it is in French) and it is just that, a large European laboratory. The CERN facility is structured something like the Los Alomos National Lab in New Mexico, or in other words, sort of like a college campus. Walking into the lab mostly you see rows of office buildings and then rows of industrial buildings stretching out over the countryside. In actuality there are several separate laboratory sites around the area. My office is in the largest of the sites, and most of the smaller sites just contain a few buildings.
CERN, being a large laboratory, has many experiments being carried out at the moment. Some of these experiments require very large machines, such as the experiment I am working on. The largest of these is the LHC, or Large Hadron Collider, which is a huge particle accelerator. From the surface, the LHC is invisible because it is buried all underground, much like a subway tunnel. This tunnel runs in a large circle that is 27km around and can be hundreds of feet underground. Inside the tunnel is a large pipe that is about 1 meter in diameter and runs around the whole length of the tunnel. By the time the pipe is finished being installed it will be 27km long and the two ends will connect. Inside the pipe is the heart of the accelerator, which I'll get into a bit later. This accelerator is one of several on the CERN campus, but is easily the largest (the next biggest is a 10th of the size or so). Hopefully in about a year, everything will be finished being built and will get turned on. Don't expect to hear from me very much after January 2008 ;)
There is another part to this experiment which is the detector used to tell what happens with the collider. There are 2 big detectors and several smaller ones located at various points around the circumference of the tunnel. The largest of these detectors is called A.T.L.A.S (A Toroidal Lhc ApparatuS) and this is what I work on. ATLAS is big, really really really big. I recently was able to go down underground and watch as it is being constructed and it is just staggeringly... big. It is shaped roughly like a cylinder that is 45m long and 25m in diameter.
The ATLAS detector is made of quite a few diff rent pieces that do different jobs and they are layered like an onion. From the inside out the pieces are the: pixel detector, SCT, TRT, 2 calorimeters, and a muon detector. One of the projects I have is working on some of the software dealing with the pixel detector. Before I can really give any more details on what I do, I should try to explain some of the physics we do.
Physics is really just the study of how the world works. However after many years of study, we know how all of the basic things in the world work more or less. This means that to study new things we have to make use of new technology to study parts of the world that were previously out of reach. For instance computers now can calculate difficult problems quickly enough that we can use them to try to understand VERY complex systems, bigger telescopes let us see the universe in much finer detail, and bigger colliders let us observe much smaller things.
Ok, that last one may take a bit of explaining, here's the analogy that gets used a lot. Imagine that there is a very dark room which you are not allowed to enter, but that there are doors on all sides. You think there might be something in the room, but you aren't sure, how can you tell? Imagine if you took a basketball and threw it into the room a lot of times. If it always sails through, you can guess there is nothing there. If it bounces back at you, then chances are, there is something it is bouncing off of. But can you tell if it is a big square block or a pyramid shaped one? Imagine you throw the ball in hundreds of times and measure where it passes through and where it bounces off, you can start to get an idea of the shape of the object. Now what if you wanted to know if the corners of the object were rounded or not? Your basketball probably couldn't tell you, but what if you used a golf ball and threw it in thousands of times? The idea is that you can only 'see' details which are bigger than what you are throwing in. Luckily for us, we don't usually have to throw balls at things to see them, we can use light. Light, however, works in almost the same way, a light particle (photon) bounces off of an object and back to your eye, billions of them at a time. Light particles are pretty darn small, small enough that it usually doesn't matter to us. However what if we want to look at things that are smaller than a photon (ok, smaller than the wavelength of the photon)? The next step is to use an electron microscope. As the name implies it uses electrons, not photons, which, have a much smaller size. This is great until you want to look at things even smaller, things much smaller than an atom. Well here comes the cool physics trick, the more energy an electron has, the smaller it (it's wavelength) is. This is one of the things quantum mechanics tells us, and if you know of a good analogy for it, let me know, otherwise you'll have to take my word for it. This is one of the reasons we build really big accelerators, to give a particle a lot of energy so we can use it to probe really small things. Really small things = the insides of atoms.
The other reason to build big accelerators is so that we can see new things. One physics equation which most people have heard of is E=mc^2. This says, energy = mass * the speed of light squared. The speed of light is just a number (and a really big one), so this says that energy equals some amount of mass. In other words you could say 10000 units of energy = 1 unit of mass. In a nuclear bomb we change a little mass into a lot of energy. At CERN we do the opposite. Particles (protons here) go through the accelerator until they are moving very fast, that is they have a lot of energy. We can speed them up going clockwise and counterclockwise around the accelerator ring. Then we can let the two protons run into each other (which is why it is sometimes called a collider). Some of the energy the protons has can be changed into mass. What sort of mass? New particles which, in a manner of speaking, come flying out of thin air and pass through our detector! Some of these particles don't live very long at all (a billionth of a second or less) before they decay. Heavy particles tend to decay into lighter particles which decay into lighter particles etc. Eventually we are left with the lightest particles which is what the universe is mostly made up of. These light particles we can study all we want, but if we want to study the heavy ones, we have to make them. Making these particles is the other big job of the LHC, and studying them is the job of ATLAS.
So what do I do here? I have two projects that I am working on. First, as I said, I am working on some of the software for the pixel detector. One of the jobs of the pixel detector is to be able to tell when a particle is traveling through it, and in what direction. To do this, it has three layers of 'pixels' each which can tell if something passes through it. Then a path can be extrapolated by drawing a line through these three points. At the moment the detector is being tested by letting cosmic rays pass through it. These are particles which come from stars and pass through the atmosphere. Usually you can't tell when they pass through the earth, but this is what our detector was designed to do. So every once in a while (once a minute or so), a cosmic ray will pass through the layers and (hopefully) be recorded. I say hopefully because the test is still being set up. The software I am writing has the job of trying to see how well all of this is working. For instance, does it measure particles passing through it, do the lines get connected correctly, is the power turned on? The goal of this is that we well be able to work out a lot of the bugs this way and things will run much more smoothly when we start for real.
The other project I am working on is with a group to try to detect one specific particle. This particle is called the 'single top quark' and is somewhat of an elusive guy. The problem with detecting it is that it is one of those particles that doesn't live very long. 0.0000000000000000000000001 seconds or so (yes I put the right number of zeros in). Because it doesn't live very long, it never makes it into our detector. This means that we can only detect the things it decays into, not the particle itself. The downside of this, is that some other particles happen to look very similar to the things this decays into. For instance, say the 'single top quark' turns into 'b', 'w', and 'v' particles when it decays. How can we tell this apart from the case when a 'b', 'w', and 'v' are created in the collision? Or what about when the 'w' is created in the collision but the 'b' and 'v' come from the decay of a 'h'? There are a lot of these cases, and we have to be able to figure out which ones really matter to us. It's not an easy thing to do. You can measure things like the angles between the 'b', 'w', and 'v', or how fast they travel and try to find differences, but that only gets rid of some of the 'pretend' single top quark signals. My work there, along with several other people is first to try to isolate the true single top events. Another thing to think about, we've never seen one, so how can we be sure we know what we are doing (this keeps grad students like me up at night)? We also work on trying to figure out how much error there is in our methods, and how likely it is that events are single top quark events. Lastly, if we can successfully do the first two, we can measure properties about the single top quark.
Practically speaking, I spend my days on the computer writing software, debugging software, reading physics papers, running my software trying to figure out how to isolate single top quarks, re-writing my software to do a better job of it, eating lunch at the cafeteria, and drinking too much espresso (which is really really good here).