The present invention relates to acousto optic modulators (known as AOMs) for adjusting a frequency of light from a light source, typically a laser. AOMs are used for laser beam modulation, frequency shifting and deflection, and are also known as Bragg cells, or acousto-optic deflectors.
AOMs typically comprise a crystal through which a RF frequency (in the MHz range) travelling pressure wave is passed, thus creating a periodic modulation of refractive index across the crystal. When a laser beam is aligned through the crystal at a nearly orthogonal angle to the RF wave, the periodic modulation acts like an optical grating causing the beams to be split into several deflection orders. Due to the traveling wave nature of the RF oscillation (at frequency F), each deflection order (m=±0, 1, 2, 3, . . . ) is shifted by mF. The amplitude of deflected beams depends on the RF power and switching times (between the RF is turned on an off), can be below a micro-second.
Typically this is achieved using a planar piezoelectric transducer arranged to convert the electrical/radio wave input into a pressure wave, and the medium or crystal used, is generally quartz. An acoustic absorbing material is positioned at the far side of the medium/crystal to absorb the pressure wave and prevent it reflecting back.
All transparent materials have a photoelastic effect to some degree, however a material with a strong photoelastic coefficient is preferable, such as quartz (fused silica) or lithium niobate or others as are known in the field.
The deflection angle is proportional to the RF frequency. So if one wanted to change the frequency, then the beam direction changes slightly. This causes the beam to be misaligned further along the optical beam path. When laser are coupled into optical fibers, they require micron-level positioning accuracy. Therefore the ‘single-pass’ AOM configuration is only suitable for fixed frequency applications. It is possible to buy integrated single-pass AOM systems which are fiber coupled.
For some uses, it is necessary to change the frequency, sometimes frequently, and for this the ‘double-pass’ AOM system is used, in which the deflected beam is passed back into the AOM along the same incident path and can be extracted with polarizing optics (see
The double pass method normally requires additional optical elements to ensure the beam is accurately reflected back along its incident path, which means that due to the size of the system it is difficult to integrate into a compact and rugged form, unlike the single-pass systems. Hence it is believed that there are no commercially available double pass AOM systems on the market in a compact form, which limits the portability of these systems.
A quarter-wave plate is provided next to a polarising beam splitter, so that the light transmitted into the AOM is rotated (e.g. +45 degrees), reflected by the mirror reversing the polarisation 180 degrees, and then returned back through the quarter wave plate which rotates it another 45 degrees (e.g. positive 45 degrees rather than negative 45 degrees since the polarisation was reversed by the mirror), lending a final total of e.g. 270 degrees. This ensures that when the light arrives back at the polarising beam splitter it passes through to the outlet of the device (rather than returning the way it came in via the inlet). The quarter waveplate can alternatively be between the AOM and the mirror (as shown in
Typically the light blocking arrangement will be a black planar element with a slit (shown with the width of the slit exaggerated) which needs to be positioned manually on a lab bench. One advantage of the double-pass approach is that the increase in frequency is twice as great compared to passing the light through the AOM a single time. However the position of the slit often has to be manually aligned, and sometimes has to be done quite accurately. Typically it is difficult to select a new frequency adjustment quickly and reliably, and in particular difficult to provide a linear change in frequency over time (i.e. to chirp the laser frequency slowly), since adjusting the slit would simply cause the device to jump between different diffraction orders with an intervening loss in the optical signal.
Furthermore in the known double-pass arrangement the slit must not be excessively narrow. An excessively narrow slit causes diffraction and prevents the slit allowing an aligned beam through, and also makes it hard to align the mirror to ensure that the returned beam also passes through the same slit. As a result, this limits how selecting the slit can be, which in turn increases the distance that it must be positioned from the AOM, which increases the size of the overall AOM device. Accordingly these prior art double-pass arrangements tend to be bulky.
It is an object of the present invention to provide an improved AOM device in which rapid, continuously varying, and/or fine adjustments, can be performed more easily or reliably, and/or which can be more compact.
The beamsplitter is only one option for guiding the light in and out (optical isolator), and an alternative is to misalign the inlet slightly so that the reflected light will not arrive back at the inlet but rather at a different location where it can then be collected via the output, as shown in
According to a first aspect of the invention, there is an AOM device as set out in claim 1.
According to a second aspect of the invention, there is an AOM device as set out in claim 2.
According to a third aspect of the invention, there is an AOM device as set out in claim 9.
According to a fourth aspect of the invention, there is provided the method of using an AOM device as set out in any one of the claims, to convert at least some of an inlet light beam with frequency F, to an output light beam with frequency F+(N×2×RF) where N is an integer and RF is the radio frequency applied to the AOM.
The quarter waveplate 3, 3′ can either be between the AOM and polarising beamsplitter, or can be between the AOM and the mirror (or in an extreme example might be incorporated within the AOM, e.g. as one of its ports, or on the surface of the mirror)—since its purpose is to ensure the light arrives back into the polarising beam splitter with the correct polarisation so as not to be redirected back out through the inlet path/i.e. towards the laser typically. Preferably it is between the AOM and the mirror 3, to avoid birefringence effects of the quarter waveplate affecting the performance of the AOM.
Any arrangement which achieves a 45 degree rotation in each direction of light travel is adequate, for example a quarter wave plate on one side of the AOM should be considered to cover equivalent approaches such as placing an 8th waveplate in that location and another 8th waveplate on the other side of the AOM, each providing a 45 degree rotation of the light in each direction. Similarly a quarter waveplate covers variants such as a waveplate offering 415 (45+360) degree rotation, and indeed 45+nx360 degrees (where n is a positive or negative integer) and also 135+nx360. Both 45 and 135 degree rotations will be effective, because by adding 135 degrees, reversing the polarisation and substracting 135 (in the opposite direction—i.e. thus adding), this arrives at a total of 270 degree rotation (which is also a 90 degree absolute rotation). Perfect precision in rotating the polarisation by 90 degrees is not essential, but the more deviation/inaccuracy there is the less efficient the polarising beam splitter (if used) will be, and for many applications high efficiency is not required. Preferably the total 90 degree rotation is achieved accurate to +/−1 degree. Accuracy to +/−10 degrees might be tolerated by some users and for the quarter waveplate to be useful it must be accurate to +/−40 degrees.
Similarly it is generally necessary that the mirror is of the conventional type that reverses the polarisation of the light, as typical mirrors do. Some specialised mirrors have been reported in the literature which do not reverse the polarisation, and these should not be used unless measures are taken to ensure that the frequency adjusted light does not wholly pass back out through the inlet of the AOM device but rather the frequency light (or at least some of it) passes out of a separate outlet.
Similarly the lens is typically obviated in most embodiments, although there could be reasons to retain some form of lens for some purpose. If a non-flat mirror is chosen for some reason this may necessitate a lens. If the mirror is concave this might not require a lens between the mirror and the AOM although adding a lens on the other side of the AOM might then be necessary to address focusing caused by the concave mirror.
Instead the control of the device is achieved by simultaneous automated adjustment of the AOM acoustic frequency, as well as automated fine adjustment of the angle of the mirror. This approach poses a significant challenge, in that to avoid the device skipping between diffraction orders and to enable a gradual change to the output light frequency, both the mirror and the AOM acoustic frequency need to be finely controlled precisely in tandem. This is much more difficult to achieve compared to simply sliding a single piece of black paper/material with a slit laterally along a guide. The angular adjustments to the mirror required will typically be very small precise changes, which is made possible by an actuator, such as a piezoelectric actuator, as opposed to the traditional approach of adjusting optical elements on a laboratory bench which was by manual adjustment of a screwthread. In a simple embodiment the mirror has an actuator at/near one side, and a pivot some distance away from it, such as at the other side, enabling fine adjustments to the angle of the mirror. Maximising the distance between the actuator and the pivot maximises how finely the mirror angle can be controlled.
The angle of the mirror, and the frequency of the AOM are jointly selected such that light leaving the AOM with a frequency offset/change 50% of that required, will leave the AOM orthogonal to the mirror, thus returning into the AOM at the same angle it left, and as a result will then be frequency adjusted a second time by the AOM and light having 100% of the desired frequency offset will then be directed parallel to the light arriving from the beam splitter, and will enter the beam splitter so as to pass to the outlet of the AOM device.
Thus by correct selection of both the mirror angle and the AOM acoustic frequency, a desired frequency adjustment can be made, and this adjustment can be finely controlled and the output light frequency could be for example be ramped up or down gradually over time if desired.
Larger frequency adjustments are also possible, as was also the case with the prior art two-pass approach, by utilising different diffraction orders from the AOM. For example if a small frequency adjustment is desired, the 0 order or 1st order diffraction can be utilised. If a large frequency adjustment is desired, the 2nd 3rd or higher order diffraction can be selected. A gradually varying frequency adjustment is possible whilst using a consistent diffraction order, but there would be a jump or break in the light beam when changing from using one diffraction order to another. Compared to the prior art approach such jumps or breaks can typically/potentially be much faster/briefer, since in typical embodiments the mirror can be automatically adjusted to the precise angle required, far faster than a human could manually reposition the slitted light blocker to the precise position required (e.g. 5 milliseconds rather than 5 seconds, depending on the actuator used).
Compared to the prior art arrangement in
Any actuator can be used, preferably one that can provide sub-milliradian (e.g. 0.01 to 0.9 milliradian) pointing accuracy of the beam. The actuator preferably has a range of tens of milli-radians (E.g. 20 to 90 milliradians). The actuator preferably has a short settling time when switching between positions (e.g. 0.01 to 10 ms).
As an example, an 80 MHz AOM, may require an angle shift of 4 milliradians (half the beam deflection angle); the angle change is linearly proportional to the frequency shift. If the mirror is fixed, any frequency shift in the AOM drive RF will cause the system to become misaligned. By adjusting the mirror angle alongside the RF frequency, the retro-reflection can remain optimised. Moreover, it is possible to swap the reflection between different diffraction orders of the AOM, thus allowing frequency jumps between zero shift (0th order), twice the drive RF frequency, or even four times the drive frequencies (with a reduction of coupling power with higher orders) or higher. This ‘tilt-shift’ AOM configuration can therefore act like a simple optical switch, or a dynamic frequency shifter.
Preferably the AOM device is mounted on a single rigid base, to reduce beam path lengths, thermal drifts and beam expansion, to form a (compact) integrated system. A MEMs micro-mirror (such as the Hamamatsu S12237-03P) is an example of an actuated mirror which could be used.
For completeness, the mirror could be actuated to pivot via more than one hinge or more generally in more than one axis, such that the x & y angle of the beams can be controlled to couple into the fiber or other optical output path. Rather than a polarization splitting cube, a polarising splitting fiber could be using.
For completeness, by ‘an orientation of the mirror arranged such that the outlet light path is displaced with respect to the inlet light path’, this means a translation in an axis perpendicular to an axis in which the returned light varies according to frequency. Generally this entails a predetermined light inlet location and predetermined light outlet location (either in-use, or defined by respective inlet/outlet optical ports) with the mirror arranged to direct light of at least one frequency from the inlet to the outlet, such that the inlet and outlet are physically separated. In an extreme case however, it can be left to the user to direct light in from one direction, and receive light at the direction reflected by the mirror With reference to
Further embodiments are set out in the claims.
A preferred embodiment of the invention will now be described, by way of example only, and with reference to the accompanying figures in which:
Turning to
As a result of passing through the quarter wave plate, having its polarisation reversed by the mirror, and passing a second time through the quarter wave plate, it's polarisation is at 90 degrees to the incoming light, and thus it passes through the polarising beam splitter to exit at frequency F+N×2×RF, where N is the diffraction order and F is the optical frequency. The RF frequency must be selected such that the desired output frequency is achieved, and the slit must then be positioned, taking into account the RF frequency, so as to select the desired diffraction order. This is difficult to do both quickly and precisely. In addition, the required lens introduces a necessary gap (focal length), which adds to the overall size of the AOM device 1. Similarly, since the slit cannot be so narrow as to cause undue diffraction of its own, and also cannot be so narrow that any misalignment of the mirror will cause the returned light to miss the slit, this means that the slit (or the lens if inboard of the slit) must be positioned some minimum distance (L) from the AOM in order to be adequately selective and not to allow the wrong frequencies of light through. Again this adds to the overall size of the AOM device 1.
The user controls the RF frequency into the AOM 4, and also the mirror angle, simultaneously, for example using any type of control circuitry 10, e.g. a computer, such as a laptop or Arduino (RTM). This may be provided separately from the AOM device 1, or may be provided integrally within the AOM device 1 (not shown). The controller takes one or more input(s) and configures the RF frequency and the mirror angle to achieve a desired output. The input(s) to the controller may for example be the input light frequency F and a desired output light frequency. This may in turn be received from other equipment or computer devices, as part of a larger optical system, or alternatively the input may be received using a user interface (not shown) such as dials or buttons.
Of course there are alternative ways to adjust a direction that a mirror reflects which are also covered, for example providing it with a transparent liquid cell in front of its reflective surface, and adjusting the angle of one of the walls of the liquid cell. Similarly it is not strictly necessary for the mirror to be flat, although in most cases another correcting optical element would be required to compensate for it not being flat. 3D retroreflectors should not be used as a substitute for a mirror as this would render the tilting of the mirror redundant, however a lenticular array of 2D retroreflectors could be arranged to have the desired effect provided that the array was oriented so that as the angle of the array was adjusted, it would selectively return light to the AOM parallel to its exit angle, based on the angle that the light had exited the AOM.
In
In
Controller 10 is shown outside the housing 11 but could optionally be provided with the other components, or integrally within the housing 11. Where the use has a computer suited to controlling an optical apparatus, the user could connect this to control the mirror and the RF input simultaneously, so either the AOM device should include a dedicated controller, or the AOM device should be arranged for its mirror and RF input to be simultaneously controlled by the user's own computer equipment.
Turning to
More generally there is provided an Acousto Optic Modulator Device that has an Acousto Optic Modulator (AOM) 4 and an actuator controlled angle adjusting mirror, and means to pass light from an input port through the AOM to the adjustable angle mirror and back through the AOM to an output port. The effect of controlling the acoustic frequency in the AOM and the angle of the mirror is to selectively perform fine adjustments to the frequency of inlet laser light, permitting controlled or sudden changes.
According to one aspect there is provided an Acousto Optic Modulator Device 1 comprising: an Acousto Optic Modulator (hereafter referred to as an AOM) providing an optical path therethrough, having a first optical port 4′ arranged at a first end thereof, and a second optical port 4″ at a second end thereof, —wherein the AOM comprises a transparent medium arranged to transmit light between the first and second optical ports 4′ 4″, having a refractive index of light that varies in response to the pressure of the medium, and comprising a pressure transducer arranged to receive a radio frequency RF and to generate acoustic waves crossing the transparent medium; —a mirror 5 arranged at the second end of the AOM arranged to reflect light from the second optical port 4″ of the AOM back into the second optical port 4″ of the AOM; and—light distinguishing means comprising either: an orientation of the mirror arranged such that the outlet light path is displaced with respect to the inlet light path, or a quarter wave plate arranged outboard of one end of the AOM, such that the polarisation of inlet light having a dominant polarisation direction, is perpendicular to that of outlet light; and Characterised in that the Acousto Optic Modulator Device 1 further comprises: a controllable actuator 8 arranged to adjust the orientation of the mirror 5; and a controller 10, arranged to simultaneously adjust both the radio frequency into the pressure transducer to control a frequency of acoustic waves crossing the transparent medium, and the controllable actuator 8 to control the orientation of the mirror.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2201108.4 | Jan 2022 | GB | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/IB2023/050418 | 1/18/2023 | WO |