Detectors 101

By JoAnne Hewett | December 14, 2006 12:45 am

While theorists routinely work with the most fundamental degrees of freedom in their calculations, the world of an experimentalist, i.e., the real world, is quite different. Experimenters must cope with particles that decay too quickly to be observed; particles that don’t exist freely by themselves but only in bound, hadronized, and fragmented states that keep showering into even more particles; and particles that cannot be detected at all. And we theorists expect them to relate the pile of stuff they observe to our fundamental degrees of freedom, and to get it right, every time. Experimenters accomplish this feat with what looks like a huge pile of semi-organized chunks of metal, liquid, gas, wire, and cable, called a detector. It’s amazing, really. I am always in awe when I visit a detector! In order to interpret LHC (or Tevatron or B-Factory) data, a theorist must have at least a rudimentary knowledge of how a detector works.

Modern collider detectors are known as 4pi, or hermetic, detectors. (Burt Richter led the team which built the first 4pi detector in the 1970’s at SLAC.) A hermetic detector is simply one which completely surrounds the collision point, i.e., in more technical language, it covers all 4pi steradians of solid angle around the collision. The shape of a collider detector is essentially cylindrical and one can think of it as a can, with the interaction point being located at the center of the can. The can has a barrel piece, which has the beampipe as its axis, and two endcap pieces which fit in snugly to the barrel region. The snug fit is incredibly important, and the degree of snugness is called hermeticity. A detector whose components do not have a snug fit is not hermetic – it would allow particles to escape undetected. If this happens, not only do you lose the particles that you should see, but you have over counted the production rate of the ones you shouldn’t see. That’s obviously a bad situation. Unfortunately, no detector is ever perfectly hermetic — they all have cracks and experimenters learn how to fold that into their data analysis.

These cylindrical detectors are a series of subcomponents, each wrapped around each other, and each performing a specific job. A general all-purpose collider detector has the following components (looking at a slice view and starting at the interaction point and going out):

  • Vertex detector: This is a very tiny component which surrounds the interaction region as closely as possible. It is basically built of layers of silicon and if a particle has a lifetime of order a picosecond (10-12 seconds) then its decay vertex (i.e., the track of the single particle splits into the 2-3 tracks of its decay products) can be observed. Bottom quarks, and sometimes charm quarks, can be identified in this manner. The vertex detector is the first component to get fried if something goes even slightly amiss with the beam.
  • Tracking Chamber: This component determines the trajectory of charged particles. The electromagnetic energy loss (via interactions with the medium in the tracking chamber) and momentum of a charged particle can be measured. It works by tracing the helix of the charged particle as it traverses the chamber in a magnetic field. The chamber is made of layers of finely segmented material, usually silicon.
  • Electromagnetic Calorimeter: High energy electrons and photons interact with the material in the ECAL and create showers of particles — this process occurs at an exponential rate allowing for the ECAL to absorb all their energy and they finally come to rest. The energy of electrons and photons is thus measured. Various materials are used (lead crystals are popular), but the calorimeter is usually transversely segmented.
  • Hadronic Calorimeter: Hadronic particles (jets of particles made from hadronized quarks and gluons), usually protons, neutrons, pions, and Kaons, interact with the material in the HCAL, creating showers, coming to a stop and thus depositing all their energy in the HCAL. This usually requires a fairly dense medium (steel scintillators are popular) and is also generally transversely segmented.
  • Muon Chambers: These are huge chunks of iron that surround the outside of the detector. Muons are heavy and relatively long-lived — thus they traverse the rest of the detector without stopping and track through to the outside chambers. Supposedly, muons are the only particle that can travel through the detector without showering and stopping in the calorimeters, but sometimes very energetic pions can make it through the hadronic calorimeter and punch through into the muon chambers. This is known as pion punch-through and gives a fake muon signal.
  • As you can tell from the above list, every detector also has a magnet, in order to track charged particles. The size and shape of the magnetic can vary quite dramatically, but it is usually quite large with state of the art magnetic field strength.

    A nice graphic illustrating these various detector subcomponents and the particles they are designed to identify is:

    The solid lines illustrate when a particle leaves a track in a detector component, and the showers depositing all their energy in the ECAL and HCAL are also shown. Neutrinos, by the way, sail smoothly through the detector unnoticed, and unbothered by all the material placed in their path.

    This is a very basic and rudimentary description of collider detectors, but one that I give on the first day of class when I teach a Collider Physics course. A nice set of lectures, written for theorists, describing detectors and collider physics in more detail can be found in hep-ph/0508097 by my good friend Tao Han at the University of Wisconsin. I admit to stealing the above graphics from these lecture notes.

    As you can see, detectors are very complicated and finely tuned instruments and my hat is off to the experimenters who make them work and give us the data!

    • Carl Brannen

      An excellent post, and a nice education to read before looking at the mysterious events seen in another type of particle detector, the Centauro events, seen in cosmic ray emulsion chambers. The events consist of an unexpected ratio of hadrons to photons, or charged pions to neutral pions, which is a violation of the expected symmetry. For more on these events, see hep-ph/0111153

    • Michael Saelim

      Awesome stuff! I’ve had some involvement in the High Resolution Array (HiRA) group at NSCL, which uses a spherical 4pi detector for beam collisions with stationary targets, and I’ve visited D0 at FermiLab. The detector was in place and behind concrete blocks for shielding, unfortunately, but I still got a sense of the enormous size of the detector – as tall as a ginormous warehouse! Kudos to the experimentalists – I can’t imagine trying to account for things in the real world.

    • Sean

      Count me in as someone who is consistently awed by actual detectors. You look at a piece of paper with a Lagrangian written on it, and then at the looming behemoth with wires sticking everywhere and people scurrying around in hardhats, and you say “This has something to do with that?

    • fh

      “Higgs Boson at the LHC” was the focus of my Experimental Physics exams, in the preparation for that I came to not only thoroughly respect but actually enjoy and appreciate experimental physics.

      I completely agree that theorists should know this stuff, most of all because it’s fun, and equally importantly, to retain a sense of the fact that the task at hand is to describe reality as probed by these machines, and the ingenuity of the people interpreting the data they spit out.

      Sean, one could argue that the even stranger observation is looking at the Lagrangian and then looking around your office and the richness of structure of reality and wondering “that has something to do with that?”. In fact we have to create highly particular and unusual configurations of matter to try to find a behaviour of reality not arising from the SM Lagrangian! 😉

    • fh

      Oh yes, JoAnne thanks for the link to the lecture notes they look like a great read!

    • ack

      Not to mention the precision electronics used to read these things out, the sophisticated hardware/software triggers to look only at the interesting stuff, the complex data acquisition/analysis systems to label, monitor, and keep track of the data, and the sprawling simulations to verify everything.

      But man, oh, man, is it fun.

    • Joseph Smidt

      What a great post! Particle detectors are amazing! I tip my hat to all people who do this line of work. The world is better off from your work. Thanks again for the post.

    • Pingback: Detector overview « Charm &c.()

    • agm

      Man, I miss studying physics.

    • EDT

      The Discovery channel has an engineering show (something with a name like “MegaBuilders”) that showed the construction and installation of the Atlas detector(s). It was VERY impressive. I think the detector weighed something on the order of 700 tons(?) and had to be lowered 30-40 feet (without touching any walls, other equipment, etc) into the chamber for installation.

    • Sourav

      Roman pots

      I was excited to learn of these devices.

    • graviton383

      Excellent job on a very necessary post! Seeing these detectors is a very memorable experience. Steven Hawking visited both ATLAS and CMS last month at CERN (before they begin to observe TeV-scale black holes when the LHC turns on!)

      EDT: The ATLAS and CMS detectors weigh on the order of 10,000 Tons and are as large as a building. What you saw on TV was the lowering of a single part of the detector down to the pit where ATLAS is being assembled. An interesting tidbit from an experimeter friend is that the DENSITY of ATLAS is such that it would float on water whereas CMS would sink!

    • Joseph Smidt

      Thanks again also for referencing hep-ph/0508097. I’ve been reading the paper and it is great. I learn so much from this blog.

    • Alejandro Rivero

      “and the degree of snugness is called hermiticity. A detector whose components do not have a snug fit is not hermetic”

      hermiticity is related to unitarity, is it?

    • Pingback: Life on the Lattice()

    • zq

      I am just curious that if this blog allow anonymous reply, why doesn’t it get lotsa spams.

    • Pingback: On particle detectors « Entertaining Research()

    • Arun M

      Thanks JoAnne, for that down to earth introduction to detectors. The article by Tao Han also made very good reading. As a particle theory student it is nice to understand how the cross sections/lifetimes etc are actually measured by the experimentalists. I look forward to more posts of this nature.

    • Jennifer Ouellette

      I have clipped and filed this particular post for future reference. It’s one of the best breakdowns I’ve encountered of how, exactly, physicists can “see” something so tiny that lasts a fraction of a second… something that non-scientists often have a very difficult time grasping/appreciating.

    • Pingback: Off-Lattice Links « Life on the Lattice()

    • Michael Saelim

      Alejandro, I believe you’re thinking of Hermitian operators. The hermeticity/hermiticity in this post is more related to the idea of a hermetic seal or hermetic calorimeter.

    • JoAnne

      Alejandro, that was a typo in the text – should have been hermeticity! Hermeticity is a term used in calorimeters to describe maximal coverage for particles under all emission angles, combined with minimal leakage. I’ve fixed the typo in the text now… Hermiticity, on the other hand, is a mathematical property of matrices and operators. The Hermitian conjugate of a matrix is equal to the complex conjugate of the transpose of the matrix. A unitary matrix is a matrix whose inverse is equal to its Hermitian conjugate.

      I suppose the theorist in me forced my fingers to type hermiticity instead of hermeticity!

    • Sean

      Could someone tell a poor cosmologist why muons don’t get stopped by the electromagnetic calorimeter like electrons do? And don’t tell me “it’s because they’re more massive,” although I suspect that has something to do with the answer. But aren’t they all relativistic in any event, so why does the mass matter? Greater momentum transfer is needed to make them non-relativistic?

      (zq– we use SpamKarma, which is an excellent spam filter. In fact we get hundreds of spam comments every day, very few of which get through. Sometimes it’s overzealous.)

    • anon.

      Sean: “it’s because they’re more massive.” In particular, the radiation length scales like mass squared. You should be able to see precisely why this happens if you try to calculate bremsstrahlung, or look up the calculation. You end up with Lorentz factors, which are factors of E/m.

    • Aaron

      Beautiful post!!! I never realized just how dense those detectors are — from your description, it sounds like they’re made mostly of solid metal! I’m also quite amused by the notion that an off-target beam can actually damage the detector! Even TeV particles have so little energy that it’s hard to see them wrecking anything except each other, but I guess a steady beam can go a long way. It’s certainly nothing I’d like to be sitting in front of! :)

      p.s. “Pion punch” would make a great name for a drink. Preferably one that scintillates! Quinine-containing beverages like tonic water and quinquina often fluoresce nicely under UV light… 😀

    • Jeremy Chapman

      Great post, and very concise. Detector physics is such an interesting area to me because I am always trying to ‘visualise’ physics. I can also attest to the intricacy of the vertex detectors as I am working on a potential upgrade for LHCb’s vertex locator. designing readout electronics that trigger and pick up data from collisions every 25ns is quite a task!

    • Sean

      Thanks, anon., that makes sense. One of those things that just never got explained to me. (I also taught a particle physics class, but we, um, didn’t do a lot about detectors.)

      Aaron– A TeV isn’t a lot of energy by macroscopic standards, but you absorb a hundred billion such particles per second and it begins to add up!

    • Pingback: The Real World | Cosmic Variance()

    • Roberto C

      Very concise and informative post. There is something that can be misleading tough, both in the text and in the picture. The hadronic particles can start their shower in the electromagnetic calorimeter, and they usually do. Just, their shower is less compact than an electromagnetic shower, and need a much deeper calorimeter to be fully contained. So the energy of an hadron is the sum of the energy deposited in the ECAL and in the HCAL. A naive reader (if any in this wonderful blog) can be lead to the conclusion that hardons do not interact in the ECAL.
      And, I would say that the muon detector is a big chunk of iron instrumented with tracking chambers.

    • Lab Lemming

      Pardon the silly question, but how do you determine how many tracks your big lump of iron actually contains? I hope you don’t have to grind down and etch the whole damn thing…

    • Pingback: LHC Detector Performs First Test of Fundamental Forces | Cosmic Variance()


    Discover's Newsletter

    Sign up to get the latest science news delivered weekly right to your inbox!

    Cosmic Variance

    Random samplings from a universe of ideas.

    See More

    Collapse bottom bar