Abstract

In coherent optics the pulse area θ is defined as the total angle the atomic state vector rotates around the electric field of the resonant driving pulse. The atomic excitation remaining after passage of the pulse is determined by this angle. For example, a π/2 pulse will put the atoms in a coherent superposition of the ground and excited states, a π pulse will put the atoms in the excited state, and a 2π pulse will take the atoms to the excited state and then back to the ground state. Another type of pulse which has generated a considerable amount of theoretical interest¹ but for which there has been only a few experimental investigations² is the 0π pulse. For this pulse, the initial part of the pulse excites the atoms and then due to a phase change of π in the electric field, the latter part of the pulse takes the atoms back to the ground state. Thus, just as in self-induced transparency, the atoms are completely in the ground state after passage of the pulse. There are two ways to produce such a 0π pulse. The first is to switch the phase of the second half of the pulse by electro-optic techniques. For this method it is difficult to get the 0π pulse area to the desired accuracy. The second method is to rely on the automatic reshaping to the 0π pulse which occurs when a weak (θ(t) ≪ 1) pulse is propagated through a resonant vapor. This method will produce the 0π pulse area to any desired degree of accuracy. This fact is due to an area theorem for these weak pulses which states that the pulse area decays exponentially with the on-resonance absorption coefficient.¹

© 1984 Optical Society of America