Sp. Shah et Sa. Rice, A test of the dependence of an optimal control field on the number of molecular degrees of freedom: HCN isomerization, J CHEM PHYS, 113(16), 2000, pp. 6536-6541
This paper reports the results of a study of the robustness of the field re
quired to generate the isomerization reaction HCN --> CNH as a function of
number of degrees of freedom and the complexity of the description of the d
ynamics. The particular reduced state representation of the control process
that is tested in this paper is the reaction path method proposed by Zhao
and Rice. We show that if the description of the system dynamics includes v
ibrational motions perpendicular to the one-dimensional reaction path and b
oth the interactions between those vibrations and between them and the reac
tion path, the fields found by the conventional computational scheme repres
ent local optima, and none of these correspond to generating a transfer of
100% of the population from the ground vibrational state of HCN to a mixtur
e of vibrational states of CNH. Moreover, it is very difficult to find fiel
ds that will efficiently transfer population from the ground vibrational st
ate of HCN to particular vibrational states of CNH. Comparing the optimized
control fields reported in this paper with those previously obtained using
simplified versions of the reaction path reduction, one finds that the com
plexity (measured by the power spectra) of optimal control fields increases
as the dynamical description includes more degrees of freedom and then the
interactions between all of the degrees of freedom. The optimal control fi
eld generated using a simpler dynamical description is not a good guide to
the optimal control field associated with a more complex dynamical descript
ion. We conclude that the reaction path method of reduction of the complexi
ty of calculation of the optimal field required to drive a particular react
ion is not likely to be useful for the design of fields with which to activ
ely control reactions of polyatomic molecules. (C) 2000 American Institute
of Physics. [S0021-9606(00)02140-1].