Isomerization Pathway Confirmed by Neutron Diffraction

Although it was clarified that the 4-cb group of I is isomerized to 1-cb through 3-cb and 2-cb, the 4-cb group of III was directly isomerized to the 1-cb group and the intermediate structures with 3-cb and 2-cb were not observed. In order to elucidate the reaction mechanism, deuterium atoms were introduced in the 4-cb group and 

The longer wavelengths around 700 nm were selected and the crystal size was 4.5 X 1.5 X 1.4 (ca. 9.5) mm was used. The crystal of III was exposed to a xenon lamp for 2 days at 273 K [25]. A small part was cut from the crystal and the structure was analyzed by X-rays. The occupancy factor of the disordered 4-cb and 1-cb groups was obtained as 0.528(6) 0.472(6). 

Possible three reaction processes in the photoisomerization from he 4-cb to 1-cb 

The third one, (c), is called alkyl turn isomerization. This process also involves no intermediates with 3-cb and 2-cb groups. The C atom of the photo-produced 4-cb radical abstracts one of the deuterium atoms bound to C, then the 4-cb radical is turned upside down. Finally, the bond between the cobalt and C atom is formed. The distance between the C and the deuterium atoms is 2.66 A, which is shorter than the snm of the van der Waals radii. This means that the two deuterium atoms in the structure of 1-cb group should be located at the C and C atoms, which is in good agreement with the observed neutron structure. Since the deuterium atom bonded to the C atom shows 100 % occupancy factor, the process of the alkyl turn should be irreversible. These results clearly indicate that the reaction proceeds in alkyl turn pathway. 

The alkyl turn isomerization is very favorable in the crystalline-state reaction. To explain the movement of the 4-cb group, the reaction cavity for 4-cb in the crystal structure before irradiation is shown in Fig. 6.22a. The reaction cavity after irradiation, which includes both of 4-cb and 1-cb, is shown in Fig. 6.22b. Since the two cavities are very similar, it is reasonable to assume that the reaction cavity does not change to a great extent during the photoisomerization. 

---