deactivating group

If an activating group makes a benzene ring react faster, a deactivating group does the opposite: it makes the ring slower and harder to substitute than plain benzene. Nitrobenzene, for instance, is so deactivated that nitrating it a second time takes hot, harsh conditions, and Friedel-Crafts reactions on it simply do not work at all.

The cause is again electron density, but in reverse. Deactivating groups pull electron density out of the ring, leaving the pi cloud poorer and less willing to attack an electrophile, and — crucially — leaving the positively charged arenium ion intermediate less stabilized and harder to form. Strong deactivators like -NO2, -C(=O)R, -SO3H, -C(triple)N, and -CF3 withdraw electrons by induction and (for those with a double bond to an electronegative atom) by resonance that drains the ring. The more electron-poor the intermediate, the higher the energy hill of the slow step, and the slower the reaction.

Just as activators pair with ortho/para-direction, strong deactivators pair with meta-direction — for the same underlying reason. A meta-directing deactivator destabilizes the arenium ion most severely when the electrophile attacks ortho or para to it (because those are the positions where the ring's positive charge would land right on the carbon bearing the electron-withdrawing group), so attack is least bad at the meta position. The big exception worth memorizing: the halogens are deactivating yet ortho/para-directing, a special case driven by the tug-of-war between their inductive withdrawal and their resonance donation.