Kinematics    Dynamics    Experimental Methods    RRKM et al.   

From the directory 'Chemical Dynamics' we select the following programs: :

Screen shot of CHD

HC + D H + CD
on the PES

Stiff DifEqs for the
formation of HBr from
H2 + Br2

LEPS package for
Transition State Theory

arrhenusn.gif (2658 Byte) Arrhenius Plot for the thermal decomposition of N2O5

Cubic Autocatalysis with
oscillatory intermediates.
Animated Belusov-Zhabo-
tinsky reaction.

Logistic growth reaction
with Bifurcations and Chaos



Software example: DetMech (new and improved 25 Oct 2003)

If you have available a time/concentration(s) table from a recently measured kinetic experiment and would like to interpret it with a mechanism then you should now download the following demo package:

DetMech Demo

Unpack the ZIP file and read README.doc or README.htm. Then type 'DetMech Datafile' and choose 'M'(echanism Search). 'Datafile' is the name of your table, whose required format will be shown after your call. If it is not compatible with this format, then the software does not treat it correctly. In this case you should quit the program and edit the table. Now it should function properly. Just follow the advice on the screen. Your essential contribution (short of an extensive expert system) consists now in proposing a mechanism to the program in the guise of a series of elementary reactions (ER) e.g.

  1. NO + NO ==> N2O2
  2. N2O2 + H2 ==> N2O + H2O
  3. N2O + H2 ==> N2 + H2O
  4. N2O2 ==> NO + NO.

Next the program generates the rate laws expressing your mechanism and finally asks you to input a guess for the rate constant of each ER. Now the process starts: If your mechanism describes the observations satifactorily the program finds an optimized set of rate constants and graphically shows the convergence of the calculated versus the measured time/concentration functions. If you don't see converging curves, then very likely your mechanism needs refinement. You can take the option of inputting a new mechanism ... and repeat this loop until you are satisfied with the representation of your data.
The program can treat mechanisms with as many as 30 Elementary Reactions connecting 40 species.

 The datafile 'Aufgabe3.dat' may serve as test. It is a data table for the above reaction with the stoichiometry

2 NO + 2 H2 ==> N2 + 2 H2O.

Call this example with 'DetMech Aufgabe3.dat'. Input the above mechanism and then give a value for each rate constant, e.g. 0.1, and then press Enter for GO!

Should you fail in the attempt at DETermining your MECHanism you can mail me here and attach your datafile. I shall try to solve your problem. The more chemical information you can give me about your reactions the more likely I shall succeed. No consultant fee is necessary, just an acknowledgment in your publication and a postcard of your town...

Here is the solution of 'Aufgabe4.dat', another example:
Notice: This Mechanism Solver is a deterministic kinetic simulator working with a system of differential equations (4th order Runge-Kutta) derived from your mechanistic model. Volume, pressure, and temperature are not tracked, i.e. assumed to be constant or of no influence. The optimum set of rate constants is found by a least squares minimum search for the deviations of the measured concentrations versus the calculated ones. For reactions with abrupt concentration changes, e.g. explosions, or large temperature excursions with sensitive rate constants, this simulator will not work. You will find an excellent Chemical Kinetics Simulator CKS Version 1.01 for free download at IBM Research Almaden. This stochastic simulator is easy to use (excellent tutorial) and usable for the most general reaction conditions. It is not capable, however, to find automatically a mechanistic fit to your data. DetMech is. As a check, it is a good idea to run a mechanism found by DetMech again through CKS. If it comes out the same then you have methodically independent information about its validity.
In order to learn how the CKS algorithm works, run MontCarl from the Chemical Kinetics package below.