sphere. It is noted that the light signal from the LD at 660 nm
passed through the fiber type linear polarizer, and its polarization
state was nearly maintained at the output of solenoid. As
shown in Fig. 5, a small gap (indicated by the arrow) between
polarization state of output signal (gray line) and linearly polarized
state (red line) was found on the sphere, which indicates
a change of polarization state of the light signal from linear to
elliptical or circular. This can be mainly attributed to induced
linear birefringence of the optical fiber by bending from the
winding of the optical fiber around the solenoid to increase an
effective length.
Regarding the Faraday rotation, as the light is reflected directly
backward, rotation is additive due to non-reciprocal nature
of the Faraday effect. The measurement of actually reflected
power at the input side with change in the magnetic field
strength would only give the idea of isolation offered by the
fiber optic device. For this, an experiment (see Fig. 3) was carried
out by using the reflection mirror, which reflected the 660
nm excitation from the CdSe QDs doped optical fiber and the
back reflection power was measured by using the 3 dB coupler
(at 660 nm). Fig. 6 shows the variation of the normalized output
power (reflected power) of the fiber with the measured Faraday
rotation angle upon varying magnetic field from 0 to 0.160 T.
The normalization was made with respect to the output power
withoutmagnetic field. The normalized output power was found
to decrease with the increase of the Faraday rotation angle and
no output power was detected at the angle of 90 , indicating an
excellent isolation property. When the Faraday rotation angle
was over 90 , the output power increased again because the polarization
state started to return back to the original state. The
reflected optical power can be simply described fromthe Malus’
law as [28]