Location of Energy Barriers. III. Effect on the Dynamics of Reactions AB + CD → AC + BD

Abstract

A previous investigation of the effect of barrier location on the reaction dynamics of reactions A + BC → AB + C (Paper I) has been extended to the four‐atom bimolecular reaction AB + CD → AC + BD. “Barrier location” is defined in terms of a 3d “diagnostic” potential‐energy surface, namely the surface for reaction by way of rectangular coplanar configurations. Surface I had a barrier displaced by about 0.3 Å into the entry valley (or “approach coordinate”), and surface II had a barrier displaced by the same amount into the exit valley (or “retreat coordinate”). On both surfaces the classical barrier height was $E_c\text{\hspace{0.17em}=\hspace{0.17em}34–35}{\mathrm{kcal\; mol}}^{\text{−1}}$ . A total of 22 batches of classical trajectories were run, across these two surfaces. The reacting molecules were free to move in three dimensions, i.e., reaction took place across the full 7d potential‐energy hypersurface. The batches differed for the most part in the magnitude and distribution of the reagent energy. The reagent energy was restricted to relative translation ${\mathrm{(E}}_T)$ , and molecular vibration ${\mathrm{[(E}}_V{)}_{\mathrm{AB}}$ and ${\mathrm{(E}}_V{)}_{\mathrm{CD}}]$ . It was found that on surface I $E_T$ was very markedly more effective than $E_V$ in promoting reaction. The converse was true on surface II. This finding parallels that for the three‐atom system A + BC → AB + C. In contrast to the three‐atom case, a reagent energy substantially in excess of the barrier height, approaching ${\text{2E}}_c$ , was required in order to obtain significant reaction. These findings applied equally when the masses of the particles were $m_A{\mathrm{\hspace{0.17em}=\hspace{0.17em}m}}_B{\mathrm{\hspace{0.17em}=\hspace{0.17em}m}}_C{\mathrm{\hspace{0.17em}=\hspace{0.17em}m}}_D$ , and when they were $m_A{\mathrm{\hspace{0.17em}=\hspace{0.17em}m}}_B{\mathrm{\hspace{0.17em}\ll \hspace{0.17em}m}}_C{\mathrm{\hspace{0.17em}=\hspace{0.17em}m}}_D$ . The equal‐mass case differed from the unequal‐mass case principally in the fact that the former gave direct reactive encounters (according to the most stringent criterion) whereas the latter gave indirect (or “complex”) ones. For both the equal‐ and the unequal‐mass cases on surface I the greater part of the reagent energy (very largely $E_T$ ) became internal excitation of the products ${\mathrm{(E}}_V{\mathrm{\prime \hspace{0.17em}+\hspace{0.17em}E}}_R\mathrm{\prime )}$ , whereas on surface II the greater part of the reagent energy (largely $E_V$ ) became product translation ${\mathrm{(E}}_T\mathrm{\prime )}$ . (This tendency for the reagent energy to become “transposed” into the complementary degrees of freedom in the products was also noted in Paper I, for the reaction A + BC). For the equal‐mass case a test was made on surface II of the effect of distributing the reagent energy so that ${\mathrm{(E}}_V{)}_{\mathrm{AB}}{\mathrm{\hspace{0.17em}\ll \hspace{0.17em}(E}}_V{)}_{\mathrm{BC}}$ ; the reactive cross section was somewhat less than for the situation ${\mathrm{(E}}_V{)}_{\mathrm{AB}}{\mathrm{\hspace{0.17em}=\hspace{0.17em}(E}}_V{)}_{\mathrm{BC}}$ used in most of this study.