Page 371 - Electrical Properties of Materials

Integrated optics 353

transmitted
beam
Optical input
beam

diffracted
beam

Fig. 13.13
photodetector
array A spectrometer relying on
acoustic beam acousto-optic interaction.

Let us say we have an optical beam incident at an angle θ to the hori-
zontal direction. If we want to deflect it by 2θ, then we need an acoustic
wave propagating in the vertical direction [see Fig. 13.12(a)] and having a
wavelength of λ ac = λ op /2n sin θ. How could we deflect the beam by a further
θ? In order to have the Bragg interaction, we need to change the acoustic
wavelength to

λ op
λ ac + λ ac = (13.11b)
2n sin(θ + θ)
and, in addition, tilt the acoustic wave (it can be done by using an appropriate
launching array) by θ/2, as shown in Fig. 13.12(b).
Our final example is a spectrum analyser. It is essentially the same device as
the beam scanner but used the other way round. The unknown input frequency
to be determined is fed into the device in the form of an acoustic wave via
an acoustic transducer. It will deflect the input optical beam by an amount
which depends on the frequency of the acoustic wave. The deflected optical
beam is then detected by an array of photodetectors (Fig. 13.13). The position
of the photodetector upon which the beam is incident will then determine the
unknown frequency.

13.7 Integrated optics
This is not unlike integrated circuits, a subject we have discussed in detail
when talking about semiconductor devices. The basic idea was to ‘integrate’,
that is to put everything on a single chip and by doing so achieve compactness,
ruggedness, economy of scale, etc. The same idea of integration (advanced
towards the end of the 1960s) can also be applied to optical circuits with all
the corresponding advantages. In principle one could have lasers, waveguides,
optical processing circuits, all on the same chip. In practice, the results have
been rather limited due to technological difficulties and, probably, inadequacy
of the scale of effort. The economic imperative which was the driving force be-
hind the integrated circuit revolution has simply not been there for their optical