Molecular Reaction Dynamics is an instructional computer program that allows students to interactively explore the collision dynamics of simple bimolecular reactions of the type H + HX (where X is F, Cl, or Br). The program is intended for students studying this topic at the level of an undergraduate physical chemistry course.

The concept of a potential energy surface (PES) is fundamental to the understanding of modern computational chemistry tools. All interactions between atoms and molecules take place on such a surface, and any chemical reaction must necessarily involve a trip or "trajectory" across some PES territory. In principle, and increasingly in practice, it is possible to predict transition states, activation energies, and product distributions by performing calculations that explore the appropriate PES.

The notion of an n-dimensional surface is quite abstract and not easily understood by students. Generally, PESs are introduced in the context of a very simple reaction, such as H + HF, with the added constraint that the atoms are confined to a collinear geometry. In such a case, the PES is a three-dimensional surface (energy vs R_H-H and R_H-F) that can be graphed and explored visually. With a little effort, students can determine the lowest-energy reaction pathway, the transition state geometry, and the activation energy.

To facilitate student exploration, the Molecular Reaction Dynamics program uses precomputed PESs for three reactions. Students may design and perform experiments in which they test various combinations of relative translational energy, vibrational state, and vibrational "phase" to determine combinations that lead to a successful reaction.

For each experiment, the program plots the interaction trajectory superimposed on a contour plot of the PES and simultaneously shows an animation of the interaction between the particles. After a few experiments it becomes evident that not all interactions having a relative translational energy greater than the activation energy result in product formation. The importance of the vibrational state of the reactant molecule and the position of the activation barrier on the graph ("early" vs "late") becomes very apparent.

Two additional modules of the program allow further experimentation with the H + HCl system by removing the collinear constraint and allowing the user to vary both the impact parameter and the angle of approach (relative to the H-Cl bond axis). The relative translational energy and vibrational state are frozen in these modules. For each experiment, the energy of the system is graphed versus time and an animation of the interaction is displayed. In the final exercise, the computer calculates a table of reaction probabilities over user-specified ranges of impact parameter and relative translational energy, then asks the student to use this information to estimate rate constants at two temperatures and determine the activation energy.

The program is useful in two ways. Sections of it may be conveniently used in a lecture setting to graphically introduce the concepts of PESs and interaction trajectories. It is also recommended as a basis for student projects or as a laboratory experiment.

While I believe the underlying simulation component of the Molecular Reaction Dynamics program is sound and very well implemented, I have some complaints about the user interface. Before the student can perform any experiments on a particular system, it is necessary to answer a series of questions. If the user inputs an incorrect answer to a question, there is no prompting or help available and it is impossible to proceed until the correct answer (within a range) is supplied. Since some of these answers are obtained from fairly complex calculations (e.g., average relative speed, energy of a harmonic oscillator), it would be nice if help were available. Another problem is that if the user backs up a page to view previous questions and answers, they are gone and the data entry task must be repeated. I suspect that some of the interface shortcomings may be due to the program's having been developed with the use of an authoring package rather than a general-purpose programming language.

In summary, I believe the Molecular Reaction Dynamics program breaks new ground and provides an effective simulation tool in a content area where it is certainly needed. It should prove useful in a standard physical chemistry course and perhaps in the early stages of a more advanced course in computational chemistry. The user interface and the introductory portion of the program leave some room for improvement, but the underlying simulation is very powerful and should prove very helpful to students seeking to master this topic. For further information about the program, consult the authors' Web site, http://www.ccc.nottingham.ac.uk/~pczrjw/mrd.html.