CLOSING REMARKS BY STUART A. RICE Department of Chemistry and James Franck Institute, University of Chicago, Chicago, Illinois 60637, U.S.A. Received 14th May, 1979 After I accepted the invitation to say a few words tying together and summarizing the various aspects of this Discussion, I searched the printed records of previous Discussions for hints as to how to do it. I found that each of my predecessors had a different view of the requirements, and that the only common element in the many approaches is the lament that it is impossible to do justice to the range of ideas exposed, and the many facets of the discussion that swirled around them, in the short time allotted for these closing remarks. The structure of this Discussion was a little different from that of earlier ones; the change has made my task easier by virtue of the existence of the Summary papers presented by Prof.Grice, Dr. Smith and Prof. Setser, who have done much of what would have been my work. In any event, I will pick out only a few points for comment. As I listened to the papers and the remarks that followed them, it seemed to me that this meeting had two broad components. One set of contributions concerned the results obtained from mature techniques, such as molecular beam studies of reaction dynamics, the use of trajectory studies to model reactions, the nature of reactions involving atomic species in well defined electronic states and the like. These tech- niques are still yielding information of enormous value and the potential for further important contributions remains great, especially as new technology, such as the use of laser induced fluorescence for detection of product states, is added to the armoury of methods applicable to particular problems.The phenomenological theory under- lying the mature techniques is very well developed, so disagreements arise only when there are discrepancies between the results of different investigators, or when the robust deductions from experiment are used to test microscopic models or to suggest microscopic models. In this Discussion one manifestation of the maturity of this component of the field was an obvious lack of contention. The remarks made following a paper were, almost exclusively, of the nature of short reports of new work akin to that in the printed paper.The second component of the work reported at this Discussion is contentious. I refer specifically to the papers and comments dealing with various aspects of multi- photon dissociation of molecules and of photoselective chemistry. In these cases there were frequent disagreements about how the system is prepared, what happens to it and how the results are to be interpreted. Clearly, there is need for the intro- duction of new concepts which will organize our understanding of these matters. At this Discussion we have heard about many aspects of reactive scattering. The several papers dealt with revival of an old but underutilized technique, the roto- accelerator, about the effects of vibrational and translational energy on reaction rates, about the nature of energy disposal in reaction products, about the comparative reactivities of alkalis and electronically excited rare gas atoms and more.The use of potential energy surfaces, both for qualitative interpretation and for quantitative364 GENERAL DISCUSSION computer simulation of trajectories, provided a central organizing theme for all these studies. I found the discussion of the influence of trapped trajectories on the reaction cross-section, by Child and Whaley, particularly interesting. This work, which illustrates the analytical considerations of Pechukas, is a welcome example of the triumph of intellect over brute force. Leaving aside the various uncertainties that still bedevil the study of multiphoton dissociation, we heard of the possibility that both fluence and intensity are important in determining the rate of reaction, which is a departure from the predictions of the available theoretical models of the process.We also heard that there are now a very few hints that carefully prepared initial states of a molecule can lead to photoselective chemistry. There has been con- siderable scepticism, based on the predictions of statistical models of unimolecular reactions, that photoselective chemistry is possible. It is important that the limits of applicability of various models of reaction dynamics be determined, and several reports at this Discussion have dealt with studies of the domain of validity of the hypothesis that energy randomization is faster than competing reaction processes, which is the basis of the RRKM model.As in other studies of this type, neither protagonists nor antagonists of the hypothesis clearly prevailed. Despite the richness and variety of the offerings at this Discussion, I found there were enormous gaps in the coverage of the subject Kinetics of State Selected Species. I believe we should have heard much more about the following: (1) Insufficient attention was devoted to the nature of the prepared state, how it evolves and how this evolution influences what we observe. Although we have learned from the theory of radiationless processes of the intimate relationships between the character of the excitation source and the state of the system prepared, and how the system behaviour sometimes cannot be well represented as a sequence in which preparation and evolution are disjoint, our consciousness concerning this has not been sufficiently raised.Much of the contention in the interpretation of multiphoton dissociation can be traced to a lack of understanding of the nature of the prepared state. In this sense, George’s analysis of laser assisted collision processes is an ex- ample of properly placing the emphasis on the joint properties of the molecule and the electromagnetic field. We must learn how to interpret, and manipulate, the dressed states representing the interaction between a molecule and some exciting field, since an understanding of these will enable us to direct the chemical processes if and when that is possible. (2) Continuing in the same vein, there was insufficient attention focused on how to generate wave packet initial states, which must be used if we hope to excite bonds, or localized regions of large molecules.Even when considering the reactions of tri- atomic molecules, excitation of a nearly stationary state that involves contributions from nuclear motion delocalized over the entire molecule is less likely to lead to photoselective chemistry than is excitation of a particular bond. (3) Throughout this Discussion very little has been said about the properties of large molecules. I believe that the prospects for selective photochemistry are most promising in the category of bimolecular reactions of excited polyatomic molecules, rather than in the variants of multiphoton induced unimolecular decomposition. The key to developing photoselective reactions will be to establish conditions under which there is efficient competition between a given reaction and energy transfer, and this should be easiest when the excitation is in the lower part of the manifold of states where the dynamics of the coupled oscillators of the molecule can be described as quasiperiodic.GENERAL DISCUSSION 365 (4) We have heard nothing at this Discussion concerning collision-induced intra- molecular energy exchange, particularly when one of the collision partners is electron- ically excited. The very limited data available indicate that the cross-sections for internal energy transfer in that case are very large, yet strong propensity rules govern the pathways of energy transfer.Indeed, it seems possible that, with clever manipula- tion of the properties of the system, we can generate selective collisional excitation of levels that cannot be photoexcited, and use the molecules so prepared as reactants.( 5 ) We have also heard almost nothing of the influence of initial rotational state on the rate of unimolecular reaction, disposition of energy amongst the products, and so on. The very limited data available have thus far not yielded up their secrets. Al- though conservation of angular momentum must influence the reaction dynamics, in the cases for which data are available no systematic pattern of dependence of the rate of the process on initial rotational state has yet been discerned. Unravelling this puzzle will surely yield rich rewards in the understanding of reaction dynamics. (6) Finally, I was surprised that we did not hear much more about the dynamics of nonlinear systems, and the transition between quasiperiodic and stochastic be- haviour in such systems.Understanding the dynamics of coupled nonlinear oscilla- tors goes to the heart of the reaction dynamics of polyatomic molecules. We need further development of the analytical and topological theories of nonlinear dynamical systems, a better treatment of the relationship between the quantum mechanical and classical treatments of these systems, more studies of models that include rotation and the coupling of vibration and rotation, analytical and numerical studies of large amplitude motion in systems of nonlinear oscillators, a better understanding of the nature of and the conditions that determine the transition from quasi-periodic to stochastic behaviour, a simplified but accurate way to describe the dynamics of polyatomic molecules, e.g., by the use of an effective Hamiltonian, and more.I have now spoken about a number of subjects not adequately addressed in this Discussion. My purpose in doing so is not to criticize the contributions of the past few days, but rather to remind all of us that a full understanding of the subtleties of reaction dynamics requires a very broad view, and that contributions to that under- standing can be made by investigations covering the entire range from simple experi- ments to arcane mathematical analyses. I will close my remarks with a story which is intended to tickle the consciences of all participants in the ongoing debate over the nature of intramolecular dynamics and the influence of that dynamics on the rate of reaction.This story concerns a renowned art historian, acclaimed all over the world for the perceptiveness and subtlety of his interpretations of iconography. Some of these interpretations could only be called devilishly clever, and all formed part of a philosophical overview of the place of iconography in the social and pyschological structure of the local population where that art form flourished. Indeed, it seemed that no new find, however strange or seemingly inconsistent with the philosophical overview he had developed, could not somehow be fitted into his world view. At a meeting of art historians, a meeting analagous to this one, he was approached by a young man who professed his admira- tion for the work of the master. How, the young man asked, did he manage to always find the clue that permitted fitting every example into one all encompassing theory? Did the search for such clues require an unusually broad background in psychology, classical studies and the like? Was there never a case that did not fit no matter what considerations were brought to bear? The art historian smiled broadly and said: You make too much of it. It’s really straightforward - in all of my interpretations I simply bend the nail until I hit it squarely on the head!