A complete description of magnetization in the rotating frame is much too lengthy for discussion here. A good description of the whole concept of the rotating frame and how a pulse effects the magnetization vector(s) is given in a very old but quite good book called "Pulse and Fourier Transform NMR" by TC Farrar and ED Becker. You can also look at "Modern NMR Techniques for Chemistry Research" by Andrew Derome and "Nuclear Magnetic Resonance" by PJ Hore. These are just a few that come to mind - I know there are many more. These books explain the whole concept of the phase of the B1 field and how it relates to the rotating frame very well. I also have a Power Point presentation that I can send you (some of it based on Farrar and Becker) that I use in teaching.

A very enlightening paper is "The Magnetic Resonance Myth of Radio Waves" by DI Hoult (Concepts in Magnetic Resonance. 1, 1 (1989)) which helps to counter some of the strange beliefs held in the Medical MRI community and sometimes by spectroscopy books aimed at organic chemists.

Although the vector description of pulse NMR assumes a circularly polarized B1 field, NMR spectrometers typically use a linearly polarized B1 field - although it is certainly possible to generate a circularly polarized field using two coils in quadrature. Any linearly polarized field can be decomposed into two counter rotating circularly polarized fields. You simply ignore the one that is rotating in the "wrong" direction at the price of throwing away half of your transmitter power. This is exactly the same as the 3 dB loss that occurs between linear and circularly porarized antennas at two ends of a communication link.