I installed the another AOM as attached picture as following procedure.

1. Put a PBS and adjusted its angle.
2. Put an AOM and connected to a driver.
3. Drove the AOM and played with STMs to increase the diffracted beam power.

Current diffraction efficiency is about 80% and still there is a room for improvement.

In addition, I worked on the double-pass AOM alignment previously installed.

1. Put a lens (f=75mm) after the AOM to collimate the beam.
2. Put a mirror and PBS to pick up the double-passed beam.
3. Adjusted the alignment and lens' position.

The power of double-passed beam is 5.76mW in contrast with 8.86mW input to AOM.
Hence the double-pass diffraction efficiency is about 65% which is reasonable value.

The next step is (quantitatively) measure the beam jitter with scanning frequency and minimize it.