Interpretation of Some Recent Literature Data

A series of crystalline arylidene oxindoles (64 and 65) has been irradiated with the aim of photodimerization [98]. It was found that the (Z)-derivatives 65 a,b were photostable, the ( ) -derivative 64 a did not photodimerize (polymer mixtures were 

From the X-ray structural data [98] we calculate the distances (in A) of inter-molecular double bond centers in possible dimer pairs of 64, and 65. They come out as follows  

These values clearly prove that the distances of potentially reactive centers are unsuitable for predictions of reactivity. 64 a (P2 /c) and 64 b (P2/n) have distances that are very close and also 65 b (P2 /c) has distances that are not out of reach for dimerization, as several examples with longer distances are reactive (Section 2.1.5). However, only crystalline 64 b undergoes photodimerization, while crystalline 64 a does not photodimerize, and crystalline 65 b is photostable. Explanation of the different reactivities of 64 a and 64 b has been attempted with a difference of 2 kcal mol-1 in cohesion energy of the packing or with a slight twist of the phenyl ring in 64 a that was claimed to be responsible for reactions only at defects [98]. 

Such minute energetic or steric differences are not at all convincing in photochemistry with a huge excess energy for deactivation. 

Clearly, the crystal packing must be considered in order to understand these experimental data. As the geometry of the expected photodimers changes consid- 

---