AAPT Workshop, Brownian Motion to Diffusion Random Walks in Introductory Physics
Mark Reeves, George Washington University



Description
Distributed files
Materials
Other curricular elements
Ponderables
Random walk simulation
Lab: Brownian Motion
Lab: diffusion in gels
Sample Class plans
FAQs

Workshop Files

FIles were distributed on a USB drive at the workshop, and the same files are found compressed in a zip file (53Mb). Once downloaded and decompressed, a list of files is given here: Unless otherwise noted the files run on either mac or the pc machines.

  1. stats2.pdf: background readings on statistics for the students
  2. Random walk simulations are found in the file, randomwalk.zip. These are java applets that simulate 1-d and 2-d random walk motion for various numbers of particles and time steps.
  3. Ponderables on statistics: Includes simple coin flipping problems, considerations of binomial distribution for equally and unequally weighted probabilities. Ideas of distributions, mean, standard deviation are developed and linked to the concepts of average and variance.
  4. Statistical kinetics (stochastic motion) videos are captured from a microscope using an inexpensive black and white camera and a frame grabber. Here a pc platform is used. Videos are captures to files using loggerpro (insert video capture) or virtualdub.
  5. Video analysis: Once movies are captured, they are analyzed with ImageJ, found on the NIH website. This is a java-base program, freely available that can analyze still and video images. Videos are imported as a sequence of .jpg images (pc) or as .mp4 files (mac). ImageJ parses the videos frame by frame and returns the centroid of the polystyrene bead. The xy coordinates of the centroid are returned, from which can be plotted particle trajectories and histograms of particle position using any standard plotting software. Before starting ImageJ, a plug-in written by us here at GW should be installed, Bead Find (included). Simply copy the bead find folder into the Analyze folder of the Plugins folder of the ImageJ installation on your computer.
    1. For mac platforms: video files (saved as .avi format) need to be converted to .mp4, using a program called Isquint (included here). Mp4 files are imported to imageJ using the file menu, import using Quicktime. Aways choose convert to grayscale.
    2. For the pc platform: videos are converted to a sequence of .jpg images using virtualdub. These sequences are imported using the file menu, import Image Sequence. Aways choose convert to grayscale.
  6. Once the video is imported, select the area of interest, and then choose the bead find routines under the plug-ins menu. Using a threshold value of 0.3 is a good choice and be sure to check the generate boundary box.
  7. Diffusion is studied by observing the spreading of a 30 microliter drop of Bromophenol Blue (BPB) in a 1% (10 mg/ml) agarose gel. Gels are poured into petri dishes and once dried, a core of material is removed from the center of the plate using a pipette. Into the hole left by the core is injected the BPB. The BPB will diffuse over a period of 3 days, and the students take pictures of the gels several times per day (every 8 hrs for the first 24 hrs, then every 12 hours). To make the images a digital camera, light box, and home-made hood are used. The pictures are downloaded and then analyzed using ImageJ. The Plot Profile function (under the Analyze menu) is used to plot gray-scale profiles for lines selected across the gels. These intensity vs. position plots are nice bell-shaped curves that can be fit to Gaussians to extract their standard deviations. Alternately, the FWHM can be measured from plots of the data. The values of standard deviation vs. time are plotted to reveal the t1/2 dependence typical of the random-walk model.

Sponsored by NSF-DUE, CCLI Program