Quantum Diaries

As many of you may know, the ATLAS detector (and for that matter CMS) is physically huge. It weighs about 7000 tons, and measures approximately 45 meters in length and 20 meters in diameter. It is the size of a small ship, one packed with sensitive silicon sensors, sophisticated electronics, and powerful on-board computers.

When the proton beams start to collide at the LHC, we will be re-creating conditions that existed about 10^-12 second after the big bang (of course, a lot of the interesting stuff had already happened in the first 10^-37 second). All the heavy particles that we will create in these collisions existed freely in the aftermath of the big bang.

Although, these collisions will be taking place in the laboratory, one can imagine a ship travelling back to the beginning of time, and sending information about what is going on there.

— Vivek Jain, Indiana University

You may have come across the concept of “missing energy” in some of the previous posts (Adam’s from June 25, and Monica’s from last year)? What is it? Is it a signature for physics or is that we can’t count properly?

Basically, the idea is the following. When two protons collide, it is actually a part of each one of them that causes the high-energy collision, while the rest of the proton sort of “falls apart” (the latter, aka “the underlying event”, produces lower energy stuff). These two parts (quarks and gluons, generically known as “partons”) have an unknown fraction of the proton energy (or momentum) along the beam direction. However, in a plane perpendicular to this high-energy proton beam these partons have negligible energy; they are essentially moving along the beam direction. Stand in front of a wall and imagine one beam coming out of the wall and the other going into it, and the collision occurs at the surface of the wall. This collision is like a bubbling cauldron (of energy) that spews out particles every which way; hence, they have a momentum component in this transverse plane as well as along the beam direction.

From the principle of energy conservation, since the initial state (two colliding partons) has zero energy in this transverse plane, the final state (a mish-mash of electrons, pions, photons, neutrons, protons, kaons, muons, etc.) should also have zero energy. If we do a vector sum of the energy of all the final state particles in this transverse plane, we should get zero. (For instance, if the collision produces two electrons that are going in opposite directions in this plane, the vector sum of their momentum will add to zero.) If this sum is non-zero, we call it “missing energy”. So what? Does this tell us anything?

Some particles do not produce any detectable signal as they travel through the detector; neutrinos are the most common type. If they have high enough energy, events containing them will have a detectable amount of missing energy. Energetic neutrinos are typically produced in the decay of a W boson or a Top quark; lower energy neutrinos are produced in the decay of Tau leptons or particles containing a Charm or a Bottom quark; even lower energy neutrinos are produced in some decays of pions and kaons. In fact, we are able to use this feature to reconstruct a Top quark.

Most garden-variety events, other than those listed above, typically have very little missing energy. But, what about new phenomena? Well, it turns out that most models of Supersymmetry (see Seth’s post from June 28 ) posit a particle that behaves very much like a neutrino as it travels through the detector, and these can be very heavy (100 to 200 times the mass of a proton). In these models, such particles are produced copiously, thus events containing them will have lots of missing energy, much more than in events mentioned above. In fact, this is one of the unique signatures of these models. Also, there are other new Physics effects for which missing energy is a good discriminant.

However, understanding missing energy is a tricky business; you have to know the dead regions of the detector as well as the noisy ones. Imagine that one part of the detector is not working. Any particle that goes into that region is not measured. If you follow the prescription outlined above, you will get that these events have missing energy; there was stuff there but we didn’t see it. Or, electronic components in one section of the detector are very noisy, so we may think that there is lot of stuff going into that region. Once again, these events will appear to have missing energy. We also have to make a lot of fine corrections to particle energies to determine missing energy accurately. But, if we do it properly, there is a huge payoff in being able to detect new phenomena! Happy hunting.

Hi everyone, This is my first post on these blogs, and I’ll start off by talking about the ATLAS detector. Let me know what you think. Hope you like it!

Vivian wrote in a previous post,

We simulate how the particles would interact in our detector. To do this we have to have a very complete implementation in software of our detector, including the positions of all the components…Even parts like the cables that bring signals from the inside of the detector out to the electronics that register the data have to be in the simulation since there is some probability that a particle will interact in the cables!

It would be bad if there was material in the detector that we didn’t know about, which threw our measurements off, or, for that matter a bug in the simulation software that did strange things with material description, leading to the interpretation of some garden variety physics effect as a new phenomenon! One can see the headlines, “Oops. Scientists retract discovery of the Higgs boson”.

We carefully account for everything that is installed, down to its weight, composition, position, etc. Remember, the ATLAS detector weighs about 7000 tons and has an extremely large number of individual components that need to be accounted for, so this is a non-trivial task. Another problem is that when you have compound materials, e.g., cables that contain plastic, wires, etc., and we have miles and miles of them snaking their way through and around the detector, it is not easy to accurately describe their properties and precisely know all of their positions. It is also possible to make a mistake while entering this information into a database, e.g., forgetting to enter some support structure or using an incorrect or approximate description, etc.

Since physicists are skeptical by nature, we want an independent way to verify the material. So, how to do this? It turns out that we can use (real) data to “X-ray’ the detector.

This “X-ray” uses a unique property of the photon (aka “the light particle”). As a high energy photon nears a nucleus in the material it is traveling through, it can convert to an electron-positron pair. This effect is known as “photon conversion”. It is the main process by which high-energy photons lose energy as they travel through matter, and the likelihood of a photon converting depends on the material, both the amount and its intrinsic properties, that it is passing through.

In order to convert at all, (a) energy conservation requires that the photon have at least as much energy as the combined mass of the electron and the positron, and, (b) a photon, being massless, has to be near a nucleus.

The likelihood of photon conversions is quantified by a property of the material called the “radiation length” (this quantity also determines how electrons lose energy as they travel through matter); among other things, this variable depends on the atomic weight and atomic number of the element. When you have a compound material, it can be hard to estimate a value for the radiation length, and conversions provide us with an independent measure.

So, when photons encounter denser material, they undergo more conversions, and if we detect these electron-positron pairs, we can get an “X-ray” of the material in the detector. And there will be plenty of high-energy photons in our data.

Using our software, we first identify electron and positron candidates and then check if they come from a common point in space; the latter step also needs sophisticated algorithms. If they do meet at a common point in space, we form a vertex (in 3 dimensions). By looking at the position and the number of these vertices, in both simulation and data, we can decide how well the former mimics the latter.

We are also working on a complementary way to map the material using pions, protons, neutrons, kaons, etc. More about that and other details in a later post!