438 C. Hansch and R. M. Mitir 



pect 2 substitution to cause some steric hindrance of resonance be- 

 tween the electrons of the ether group and those of the ring, with 

 the result that the —I effect of the oxygen would be more important 

 and activation should result. Although 2 substitution almost always 

 results in an increase in activity, the effect is more apparent in the 

 di- and tri-halo derivatives than with the mono. A very important 

 consequence of the two-point attachment hypothesis, especially with 

 the phenoxyacetic acids, is the nature of the spacial arrangement of 

 the side chain with respect to the ring at the time of consummation 

 of reaction at the second point (ring position). If the side chain is 

 held so that electrons from the oxygen can effectively overlap with 

 those of the aromatic ring, the electron density on the ring would be 

 very much greater than if oxygen were twisted so as to prevent this 

 overlap. It would seem most reasonable to assume that at the time 

 of reaction at the second point, the side chain would usually be 

 twisted so that only weak overlap could occur. The great sensitivity 

 of the 2 position to steric factors is indicated by the fact that activity 

 drops to zero in going from methyl or iodine to ethyl. When ethyl is 

 in the 2 position, even introducing a chlorine atom in the 4 posi- 

 tion does not restore activity. 2-Ethyl-4-chlorophenoxyacetic acid is 

 completely inert. 



A consideration of structure and activity in the phenoxyacetic 

 acids is not complete without some consideration of the disubstituted 

 derivatives. All of the 2,4-halo derivatives with the exception of io- 

 dine are quite active (35). The dramatic lack of activity of 3,5-di- 

 chlorophenoxyacetic acid was first shown by Leaper and Bishop (20) 

 and then confirmed by others to hold for the 3,5-dimethyl derivative 

 as well (26, 35). That this effect is not due to either steric or H/L 

 factors is clearly shown by the discovery that 2,3,4,5-tetrachloro- 

 phenoxyacetic acid is quite active. The inactivity of the pentachloro 

 derivative again points up the importance of an open ortho position 

 (35). Both of these inactive 3,5-substituted molecules have groups 

 which would relay electrons to the ortJio positions via -j-M effects. 

 Moreover, since there is no group in either ortho position, the oxy- 

 gen atom of the ether linkage is able to exert maximum effect on the 

 ortho position tiirough -^M action. The combined effect of three 

 groups directing electrons to the ortho and para positions makes at- 

 tack at these positions by an electron-rich reagent much less likely. 

 Introduction of a chlorine atom in the 2 position should help lower 

 the electron density at the ortlio positions by providing a — I effect 

 and some steric hindrance so that conjugation of the ether oxygen 

 with the ring is not as effective. This is borne out by the fact that 

 2,3,5-trichlorophenoxyacetic acid is a moderately strong compound 

 (35). Adding another chlorine atom Aviih its — I effect to the 4 posi- 



