Molecular Beam Research
Summer 1997
A Summary Of Our Summer
Members of the Molecular Beam Spectroscopy Research team in the summer of
1996 were Marc Welge, Matthew Feig, Mike Etten, Thomas Hoft, John McKinzie, and
Nicole London.
Each of us has written a small paragraph below summarizing our
contributions to the project.
Matthew Feig
Because I also worked on the molbeam project last summer, I started out
the summer knowing a lot about it and was able to start working on
specific tasks right away. Almost the entire summer, I worked alongside
Marc and Mike, and we spent most of our time modifying computer programs
written in the programming language C. For the first seven to eight weeks
of the summer, we were working on simp, the program used to fit the
lineshape obtained from the spectrometer. We started with some minor,
rather superficial changes to the program, changes which made simp easier
to use and harder to crash. After that we took on the task of making simp
able to handle and fit lines from different isotopes simultaneously, a
much more fundamental, significant modification than anything we had
attempted before. So we were really happy (and more than a little
surprised) when the new version of simp started working almost right away
with hardly any glitches. Marc and I then spent the rest of the summer
fiddling around with beam, the program which uses the frequencies of
observed transitions to fit for the molecular constants. Our goal was to
adjust beam's diagonalization routine so that it would deal with the
entire hamiltonian matrix at once, instead of the individual submatrices
separately. We made a lot of headway, but ran out of time at the end of
the summer before we got it to work properly. Oh well!
Marc Welge
Returning for my second tour with the Molecular Beam Group, I was able to
gain greater insight into the principles governing molecular beam
spectroscopy than in my previous research experience. Last year I joined
the group four weeks late, missing out on a great deal of background
material. This summer, in addition to lectures and numerous articles, the
nature of our C-programming goals enabled us to delve deeper into the
theoretical and mathematical underpinnings of our project.
Much of my summer was spent modifying/debugging simp (the program
used to fit data from the beam) with Matt and Mike. Our crowning
achievement was isimp, the less-frequently-crashin',
multiple-isotope-fittin' version of simp. During the last few weeks, Matt
and I attempted to modify beam, a program responsible for calculating
molecular constants and generating predicted frequencies of transitions.
We altered the manner in which the energies (which appear as eigenvalues)
are calculated, but, unfortunately, the results which the program returns
are not credible.
Mike Etten
Much of the early part of my summer was spent gaining a basic
understanding of programming with C and some of the important routines we
use with the Molecular Beam. For several weeks, I accompanied Marc
and Matt in their efforts to improve the sophistication and flow of the simp
program, which analyzes the data produced by beam. The result was a
much more user-friendly, crash-resistant program which eliminates
much of the hassle in fitting data files. The majority of my
remaining weeks were spent utilizing the simp program and frequency
predictions to find and fit any new lines for our Rubidium Chloride
data. This was first done with the basic simp program, but during the
last two weeks I used the newly developed isimp (thank you, Marc
and Matt) to fit multiple isotopes at the same time. The results will
hopefully provide some new insights into previously confusing data
files, and demonstrate some interesting transition overlaps as well.
Thomas Hoft
Aside from learning all about Molecular Beam Electric Resonance
Spectroscopy, I spent the summer making changes to the program that runs
the beam itself, and analyzing data for Li7I. Basically, the changes to
the beam program were cosmetic, and made it a little more useful. The
graph of the data now automatically rescales, and the reading in of what
data to take works more smoothly. Also we added a routine that may
someday aid in aligning the beam and the detector filament. In terms of
Lithium seven Iodide, we gathered data on the J=1 lines, fitted old and
new data to get transition frequencies, and improved the fit of the
nuclear constants. Also I spent a lot of time with the apparatus, waiting
for the vacuum pumps to pump down.
John McKinzie
At the beginning of the summer, Dr. Olson informed the group that he had
an older version of the Visual Basic programming language in his
possesion. He asked if anyone would like to try to write the beginnings
of what would be the Visual Basic counterpart to, "simp," our data
viewing and fitting program. Having no programming experience prior to
this summer, it seemed a wise choice for me to take this project, so
others, with C programming experience, could concentrate on C programs.
After the first month, the initial stage of the program was done. At this
point, it would read and graph our datafiles, zoom in and out, read a
scan file and allow your to cycle through all of the data files taken.
the next step was to add the simplex method of fitting a funtion to data.
For the next four weeks, I worked on a experimental version of the
simplex method, just to get an idea of how it works and to make sure it
would work with Visual Basic. I did work and we decided to upgrade
Visual Basic to the newest edition, VB5.0. Now that the is done, we are
now starting to implement the simplex into my original plotting program.
I may continue to work on this program throughout the year in my spare
time.
Nicole London
Back to the Molecular Beam Homepage.