Chapter 3 Acousto-Optics of the textbook Crystal Optics – Properties and Applications by Ashim Kumar Bain says the following:

In general, acousto-optic effects are based on the change in the refractive index of a medium caused by the mechanical strain produced by an acoustic wave. Since the mechanical strain varies periodically in the acoustic wave, the refractive index of the medium also varies periodically leading to a refractive index grating. When a light beam is incident on such a refractive index grating, diffraction takes place and this produces either multiple-order diffraction or only single-order diffraction. The former is referred to as Raman-Nath diffraction and is usually observed at low acoustic frequencies. The latter is analogous to Bragg diffraction of X-rays in crystals and is referred to as Bragg diffraction. It is usually observed at high acoustic frequencies.

It is said that the mechanical strain causes a change in the refractive index. Since the refractive index is changing, that means that the position/direction of the beam exiting the material will change (by the angle of refraction). Furthermore, I know that acousto-optic modulators are used for shifting the frequency of a laser beam. But doesn't this mean that the use of an acousto-optic modulator for the purpose of shifting the frequency of a laser beam will also mean that your laser beam will be changing position/direction, since both effects are due to the changing refractive index? And so, if one is using an acousto-optic modulator for the sole purpose of changing the frequency of a laser beam, doesn't that mean they would also have to deal with the (potentially undesirable) effect of having the beam change position/direction? In other words, doesn't this create other problems? And yet people do indeed use acousto-optic modulators when they only want the frequency-shifting effect. Is this because the effect of the changing direction/position of the beam is negligible/insignificant?

  • $\begingroup$ a quick comment to add to the answer that makes this simple to understand: the deviation angle is proportional to the frequency shift. Just like if your AOM would be a grating where you can control the lines/distance, the diffraction angle changes depending on the frequency. $\endgroup$ Jun 1, 2021 at 12:18

1 Answer 1


You are correct, the angle of the transmitted beam depends on the order of diffraction and the frequency shifted. The effect can be significant, for example, we use AOMs in our lab to shift the frequency of a beam to image $^{87}$Rb atoms, we see a sudden drop in intensity after some change in frequency due to the shifted beam not coupling into the fiber optic cable.

However, the change in position of the beam can be canceled by passing the beam back through the same AOM, this is called a double-pass. If you are only interested in setting your beam to a single frequency where it is unlikely to shift too much, then a single pass is sufficient. A double-pass can be used when there is a need to scan over large frequency ranges, such as the cooling transition frequency in magneto-optical trapping setups. The double-pass also has the effect of shifting the frequency by twice the single-pass value.

See here for setting up a double pass aom

  • $\begingroup$ Thanks for the answer and the double-pass link. I'm not sure if you can clarify this for me, but since you seem to be knowledgeable on this matter, I'll ask anyway. Let's say I want to build a laser vibrometer with a 1 MHz center frequency (that is, the vibrations I am measuring in the target object have a frequency of 1 MHz). To this end, let's say I use a 40 MHz AOM and a, say, 45 MHz photodiode. Do I still need to set up a double-pass system, or is the beam deviation not a problem, since I only want a single frequency? $\endgroup$ May 31, 2021 at 18:48
  • $\begingroup$ If you know the frequency shift you need to set the aom to, set the aom to that frequency via the vco, then align your setup after. $\endgroup$
    – jamie1989
    May 31, 2021 at 18:52
  • $\begingroup$ You're referring to en.wikipedia.org/wiki/Voltage-controlled_oscillator ? So I effectively just change alignment of my laser system to account for the deviation of the beam? $\endgroup$ May 31, 2021 at 18:54
  • $\begingroup$ Yes, that is what I am referring to. Yes, you set your aom to the frequency shift you want, then align your setup 'down beam' of the aom to account for the deviation in the beam. $\endgroup$
    – jamie1989
    May 31, 2021 at 18:56
  • 1
    $\begingroup$ No worries, hope the experiment goes well. $\endgroup$
    – jamie1989
    May 31, 2021 at 19:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.