Abstract: |
I will start from the initial ideas and proceed through the development of the theory, hardware,
optics, data acquisition systems, and software necessary to instantiate the ideas and methods.
The time derivate method is based on the simple idea that the derivative of a constant is equal
to zero; and therefore, taking the time derivative, or time difference, of a signal will remove all
time invariant contributions to the signal. The resulting signal reflects the time varying
concentration distribution that contains the information in which we are interested. The
largest time invariant contribution to the signal is the constant background that is a function of
radius only, resulting from inhomogeneities in various components of the optical systems along
with dirt, oil, and fingerprints. I will describe the development of algorithms for model
dependent curve fitting to sedimentation velocity data employing numerical solutions to the
Lamm equation for both interacting and non-interacting systems These algorithms are based on
the methods of RenĂ© Cohen, and of Jean-Michel Claverie, as implemented by Peter Todd and
Rudy Haschemeyer in their non-linear least squares curve fitting procedures, and on methods
developed by David Cox in the 1980s.
Since the late 1990s, when these curve fitting techniques became widely accessible because of
the availability more capable desktop computing power, a software package called SEDANAL
was developed by me and Peter Sherwood. It included both time derivative, model
independent analysis (e.g. DCDT and WDA), and non-linear curve fitting to time difference data
of interacting and non-interacting systems. These techniques continue to be developed and
recently, over the last ten years, have been directed at highly non-ideal systems such as
sedimentation in serum--in collaboration with Jack Correia--and the development of techniques
for multi-wavelength analysis of complicated hetero-associating systems and non-interacting
mixtures from which both hydrodynamic and spectral information can be derived. The multi-
wavelength analytical techniques have been developed in collaboration with Johannes Walter,
Max Uttinger, and Simon Wawra in Prof. Wolfgang Peukert's group in Erlangen, Kristian
Schilling and Frank Krause at Nanolytics, in Potsdam, and Joe Pearson with Helmut Colfen's
group in Konstanz. |