Cosmic Ray Muon Detectors
Particle Physics Using Nature's Accelerator

Somewhere out there is a list of "10 Things a Physics Teacher is Least Likely to Say." If one were to find this list, it would have on it such gems as #7. Let's challenge the PE Dept to a game of rugby and #4. I don't care if you understand the concept, just give me the correct answer to 12 sig figs.

Finally, you'd get down to the biggie, the thing physics teachers never say:
#1. Let's do a particle physics lab right here at Podunk Corners High!
The traditional reasons for this are that everyone knows that particle physics is only done with Vastly Expensive and Complicated Equipment run by casts of thousands of Highly Qualified Scientists and that particle physics is Difficult and Arcane.

Yet that perception is changing . . . and the list may need updating. After all, the technical complexity of the space shuttle doesn't stop us from building all kinds of model rockets. And with particle physics, we can go one better. After all, a model rocket can't reach outer space. Yet we can build real instruments which actually detect and can even make measurements of actual elementary particles. Those particles are cosmic ray muons . . . and detecting them is not as hard as you think.

Your mission, should you decide to accept it, is to plan a workshop activity for physics teachers in which they

construct, commission, operate, and experiment with

cosmic ray detectors. But how . . . ?


Step 1: Construction
To construct a cosmic ray detector, first you need a design. We have several designs from which to choose. Each has strengths and weaknesses and each is better suited for some kinds of experiments than the others. Below is a rundown of the main detector types we are aware of, what they can be used for, their relative advantages, and where they were developed. From there you'll find a "LEARN MORE" link which goes to a more complete report on each detector and how to construct it.

The Berkeley Detector

Best Use: Coincidence muon counts
Advantages: Portability, reliability, ease of use
Developed at: Lawrence Berkeley National Laboratory



LEARN MORE

The Fermilab Detector

Best Use: Coincidence muon counts, muon lifetime
Advantages: Versatility, can use surplus parts
Developed at: Fermi National Accelerator Laboratory



LEARN MORE

The Penn Detector

Best Use: Muon counts
Advantages: Low cost, instructive design
Developed at: University of Pennsylvania



LEARN MORE

Commercial Geiger Detector

Best Use: Coincidence muon counts
Advantages: Low cost, portability
Developed at: Aware Electronics



LEARN MORE

Step 2: Commissioning
Commissioning is the important last step before a detector begins operation. To commission a detector, physicists make sure everything works as it ought to and calibrate it so that the data generated is comparable with that from other detectors worldwide. There are several cosmic ray detectors that you might work with and each has its own specifications and uses. Nevertheless, most cosmic ray detectors have certain traits in common which you ought to check.

The following items are important to good detector operation:

Good Connections: Make sure you've made all required electrical connections with good contact. If there is doubt about some of your connections, check for continuity with a multimeter.
Power: Make sure your power supply is running at the correct voltage.
All Systems Go: Is everything "on" that is supposed to be?
Configuration: See to it that everything is in place. For example, for a coincidence detector, one counter should be on top of the other. (The coincidence configuration eliminates electronic noise generated by each counter operating separately.)
Integrity: Nothing should be getting in or out that oughtn't. An important example, for scintillating detectors, is light-tightness. The paddles and PMTs should get zero ambient light: the only light inside the scintillator should come from scintillations due to charged particles. If one or more counters is reading rapid counts off the scale, it is probably not light-tight and needs to be recovered.
Counts Real Particles: If you can get a known radiation source (example: a thorium lamp mantle), bring it in the vicinity of your operating detector to see if it gives an increase in counts. Vary the distance of the source from the detector: does the count fall off as 1/r2?
Background: You want to see cosmic ray muons, not background radiation. Find a way to check background. One way, with a coincidence detector, is to separate the counters, putting them a few feet apart horizontally. Occasionally, two different particles will be picked up by the detector—one to each paddle—and be counted as a coincidence if they are close together in time. These are called "accidentals." The rate of accidentals should be subtracted from your count rate.
Comparison with a Standard: Find out how your detector compares with another of similar characterics. Compare your cosmic ray flux with published values. To get an idea of how many counts your detector should get per unit time, consult the Particle Data Book or the SLAC Cosmic Ray page.
Compare Uncertainties with Statistical Expectations: Let's say you take a count for a time interval and get 205 counts. You repeat the experiment and get 198 counts. Is this within the range of expected error? Well, a rough estimate of statistical uncertainty is the square root of the number of counts: so if you get 205 the first time, 2051/2 = 14.3, so our second count of 198 is well within this expected uncertainty.

>
Accidentals can be a problem.


Step 3: Operation
So you have your detector. It works, and it is commissioned. Turn it on!

The first thing you'll see, either on an LED display or a computer screen (or both) is numbers registering. You are counting particles passing through your detector—some of which are cosmic ray muons. If you have a two-paddle counter, you can set your electronics to record only coincident signals. You've filtered out noise. The remaining counts are mostly muons!

If you leave this device running in the classroom, it will no doubt arouse the curiosity of some of your students. It is the perfect "teachable moment" for them to find out about the swarms of charged particles which pass through them every day.

The next step is to collect data with your colleagues or with your students. There are specific recommendations for experiments below, but the first thing you should do is "play"with your detector. Try to find out what causes the muon count to go up or down and by how much. Get used to taking data and running the equipment.

Step 4: Experimentation
You have a working cosmic ray detector—or group of them. It's time to do some physics!

You will find four suggestions below. These are not, however, step-by-step instructions (except in the most rudimentary form). They are intended, rather, to be sources of ideas with references for further study and planning. You will find the ideas are not mutually exclusive, and you might be able to devise different experiments as well as new combinations of these four.

Good luck and good muon hunting!

The Lifetime of the Muon

Purpose: Measure the lifetime of the muon.
Equipment Needed: Single or double-paddle detector with thick scintillator and fast electronics.
Method in Brief:
  1. Measure time intervals of "double hits" in scintillator.
  2. Graph time interval bins vs. frequency in a spreadsheet.
  3. Fit to exponential decay curve.
  4. Derive lifetime from decay constant.
References:

Muon Flux Experiments

Purpose: Measure how muon flux is affected by an independent variable such as altitude.
Equipment Needed: Counting detector, portable if possible.
Method in Brief:
  1. Measure the number of hits (minus background) in some time interval; try to have a long enough interval that you have at least 100 hits.
  2. Change your independent variable (e.g., altitude).
  3. Repeat steps 1 and 2 enough to get a statistically significant sample.
  4. Graph muon flux as a function of your independent variable.
  5. Analyze the data.
References:
  • SLAC Cosmic Ray Data Center Guided Tour
    A description of the cosmic ray detector at the Stanford Linear Accelerator Center as well as several experiments that can be done with a variable affecting muon flux. It is designed to go with online data from Stanford, but several of these experiments can be done with your detector.
  • Snowmass 2001
    A description of the measurements QuarkNet teachers have taken of muon flux as a function of altitude using the Berkeley detector.
  • Investigation Data: University of Rochester Summer 2001
    Different muon counting experiments done by the QuarkNet teachers in the University of Rochester Associate Teacher Institute.

Measurement of g-2

Purpose: Measure the magnetic moment related parameter g-2 of the muon.
Equipment Needed: One or two paddle detector with fast electronics, a large electromagnet
Method in Brief:
  1. Perform the muon lifetime experiment with the electromagnet off.
  2. Turn on the electromagnet.
  3. Repeat the muon lifetime measurement.
  4. Repeat for different values of the magnetic field, if possible.
  5. Graph bin interval vs. frequency.
  6. Fit to 5-variable exponential/cosine function, derive value of g-2.
References:
  • Amsler, C., The determination of the muon magnetic moment from cosmic rays, American Journal of Physics, Vol. 42, p. 1067, 1974.
    Dig a bit to find this article that has information on how this experiment has been done.
  • E281 Muon (g-2) Home Page
    The official webpage of the muon g-2 experiment E-281 at Brookhaven National Laboratory. It has an explanation of the muon ring at BNL for measurement of g-2, theoretical underpinnings, the importance of these results for the Standard Model and more.

Cosmic Ray Air Showers

Purpose: Determine frequency, patterns, directions and characteristics of cosmic ray air showers.
Equipment Needed: Counting detector, GPS, computer with Internet connection
Method in Brief:
  1. Mount cosmic ray detector on roof of a building; keep it in constant operation.
  2. Feed data from CRD and GPS to computer.
  3. Feed cosmic ray, position and time data to a large network of similar systems.
  4. Central computer in network searches for cosmic ray showers in data, records appropriate measurements.
References:
  • CROP
    CROP is the Cosmic Ray Observatory Project which uses a large array of cosmic ray detectors at high schools to search for air shower events. One of the first such projects in the United States.
  • NALTA
    North American Large area Time coincidence Arrays: A single reference for most of the air shower arrays in North America. It has links to arrays such as ALTA (University of Alberta), WALTA (University of Washington), CHICOS (Los Angeles area) and SALTA (Snowmass, CO), as well as CROP.
  • High Schools Join in the Search for the Most Energetic Particles in the Universe
    Article by CROP Director Greg Snow in Fermi News on large air shower arrays (pdf file).

Project Contact: Ken Cecire - ken.cecire@hamptonu.edu
Web Maintainer: qnet-webmaster@fnal.gov
Last Update: March 22, 2002