Final Term Paper
To demonstrate your in-depth understanding of the frontier theories, experimental techniques and statistical analysis methods, as well as to stimulate your creativity in design and planning, each student in 8.811 is required to write a term research paper, and furthermore to present it, near the end of the semester, to the class for 35 minutes, including discussions. The written reports are officially due in my office. If you have extraordinary difficulty, you are urged to arrange with me individually as soon as possible to extend the deadline. I will be available for appointments to help you with your research papers throughout the semester.
By the end of October, you should have selected a physics topic for your final term paper on "How to establish a new important piece of physics." Possible topics includes but not limited to searching for the Standard or non-standard Higgs, lepton number violation, Neutrino masses and oscillations, heavy quark physics, SUSY, Technical color, sub-quark structures, running coupling constants, compactness, quark-gluon plasma, gravitational waves, etc. On paper, you should aim at improving the current experimental limits by at least one order of magnitude. Do start discussing with me early in the semester and do it often on how to carry out such an experiment. In general, you should avoid using the same topics as in 8.881.
The following sections, as a minimum, should be included in your final written term papers:
- The significance of this proposed physics;
- Its unique signature; and
- The theoretical resolution of this signature.
- What has been achieved so far?
- How would you propose to improve the current limits?
- The source,
- The detector and
- The measurable and the selection method.
- Experimental Signal: Define the experimental signal of your proposed piece of physics.
- Intrinsic Resolution: Estimate the intrinsic resolution of this signal by including relevant physical effects of the outside world to the theoretical resolution defined in 3rd bullet point under section 1, e.g. beam spread in phase space, nucleon motion inside a nucleus, multiple scattering, bremstrahlung, pair production, scattering, etc. both in the target and in your detector. This is the resolution you would obtain with a super detector with perfect spatial and energy resolutions.
- Detector Resolution: Estimate the detector resolution of this signal by including relevant physical effects, e.g. magnetic field bending, electric field accelerating, emissions, spatial digitization and energy resolution, etc. using your proposed detector.
- Detector Design: In general, you should design your detector such that the detector resolution is slightly better but no more than a factor of two better than the intrinsic resolution of this signal. You should avoid over design your detector. For example, if you fill the space using pixel detectors with a few micrometer resolutions, the number of channels will be very large. If you read them out in sequence, your event rate will be very low, limited by the readout time required. If you read them out in parallel, the effects of the multiple scattering, bremstrahlung, etc in both the detector and their readout cables might spoil your resolution, even though you have very good intrinsic detector spatial and energy resolutions.
- Experimental Resolution: Combine the intrinsic resolution, and the detector resolution, smeared out by the statistical fluctuation of background events defined below, to obtain the experimental resolution.
- Background: Define the background within the experimental resolution - how to suppress the background? The background will contribute to the uncertainty in your signal and may spoil your selection of the signal, due to:
- The statistical fluctuation of the signal and background events from bin to bin;
- The systematic error due to the uncertainty in the magnitude and the distribution shape of the background events.
- Signal to Noise Ratio: Compute the signal to noise ratio within the experimental resolution to show what you propose is plausible.
- Significance of Results: Error (statistical) analysis and discussions.
Example of Classical Experimental Design
Outline of Classical Experiment 1: Test QED in the Time-like Region (designed by German physicist Yost in 1966) (PDF)
Example of Discussions with Students before the Paper (PDF)
References for the Research Paper (PDF)
Sample Final Projects
Standard Model Higgs Detection and Measurements at a Linear Collider (PDF - 2.5 MB) (Courtesy of Georgios Choudalakis. Used with permission.)
Search for a Z' at an e+e- Collider (PDF) (Courtesy of Thomas Walker. Used with permission.) Comments by Prof. Chen after the paper was turned in (PDF)
Search for Neutrino Tau in the Long Baseline Appearance Experiment (PDF) (Courtesy of Feng Zhou. Used with permission.)
Supersymmetry Detection and Precision Measurements at a Linear Collider (PDF) (Courtesy of Alan Hoffman. Used with permission.)
Proton Decay Search (PDF) (Courtesy of Eric Fitzgerald. Used with permission.)