Patent Application
20170230132
Embodiments of the present invention relate to methods and apparatus for modifying the phase of a beam of light for example for routing and/or switching beams of light and adaptive optics. In particular, embodiments of the present invention relate to polarisation insensitive phase modification of beams of light.
Phase modulation of coherent light allows high-efficiency holographic image projection and beam steering; the latter applications include Optically Transparent Switches for optical networks, for Add-Drop Multiplexers for Wavelength Division Multiplexing (WDM) Telecomms, and TV multicast. Incident illumination from the incoming fibres can be randomly or variably polarised or polarisation multiplexed, and the overall rotation of polarisation of the original signal into the fibre may vary with time or day, temperature, mechanical stress on the fibre, etc. To deflect and route these signals without unacceptable losses or continuous adjustment polarisation insensitive methods are needed.
One method of routing signals uses a Spatial Light Modulator (SLM) which instead of modulating luminance modulates the phase of the reflected light in the relative range 0 to 2π, and presents blazed gratings or suitable holograms to steer the incoming signals to different output ports, e.g. using a Liquid-Crystal-on-Silicon (LCOS) backplane to display the phase hologram. The Liquid Crystal (LC) material may be a nematic material, generating an analogue blazed grating (in which case only a single linearly polarised component of the signal is modulated), or a suitable ferroelectric LC which can be polarisation insensitive but has the disadvantage that only binary phase gratings or holograms can be formed, resulting in an extra 3 dB routing loss. Background prior art can be found in U.S. Pat. No. 5,319,492 and JP2002/357802A.
In general it is desirable to suppress unwanted reflections in such devices; for some applications such as telecoms this is particularly important to suppress unwanted crosstalk. Background prior art relating to anti-reflection structures can be found in: U.S. Pat. No. 7,542,197; GB2,430,048A; US2012/0057235; and WO2012/123713.
Random polarisation can be accommodated by splitting the incoming signal into two orthogonal polarisation streams, routing each separately using suitably oriented nematic SLMs, and recombining them (with additional losses and a requirement for careful path length balancing), or by using a technique such as an internal quarter-wave plate.
Light travelling through a medium can be disrupted by variations in refractive index—e.g. in the atmosphere due to turbulence giving pressure changes or convection caused by temperature changes. For any object viewed in the far field, this alters the shape of the plane wave across the entry optics, and limits the resolution of the optics below the theoretical limit. E.g. for an astronomical telescope this gives a blurred jittering image of a star which should be a point source. For planetary or surveillance images this also gives instantaneous spatial distortion.
Adaptive optics can partially compensate for the disturbance, usually by using a deformable mirror with an array of electro-mechanical actuators. These can correct the wavefront deformation of a few microns and partially restore the wavefront.
According to a first aspect of the present invention, there is provided an optical routing device comprising a substrate; a plurality of reflective pixel electrodes disposed on the substrate; a transparent layer comprising at least one counter electrode; and a variable refractive index layer disposed between the plurality of transparent pixel electrodes and the transparent layer. The variable refractive index layer comprises a material having a first, isotropic refractive index under no applied electric field and a second, different refractive index under an applied electric field wherein the second, different refractive index is isotropic perpendicular to the direction of the applied electric field.
Embodiments of the present invention may be used to switch an optical signal between different outputs by applying spatial patterns of voltages between the pixel electrodes and the counter electrode. The different spatial patterns of voltages cause the variable refractive index layer to provide different diffraction gratings. The diffraction gratings may be configured to direct the input signal to different output channels or fibres.
Embodiments of the present invention have the advantage that since the refractive index of the variable refractive index layer is isotropic perpendicular to the direction of the applied electric field and because the applied electric field is applied substantially in the direction that the input signal is incident on the device, the routing device is polarisation independent.
In embodiments of the present invention, the transparent layer comprises a plurality of counter electrodes.
In embodiments of the present invention the pixel electrodes are arranged in rows and the plurality of counter electrodes each oppose a row of pixel electrodes. The pixel electrodes may be substantially rectangular and have a smaller dimension along the axis of the rows than perpendicular to the axis of the rows.
According to an embodiment of the present invention, the variable refractive index layer comprises a blue phase liquid crystal material.
According to a second aspect of the present invention, there is provided an optical telecommunications polarisation-insensitive beam switching/routing device. The device comprises: a backplane comprising drive electronics configured to drive a plurality of optically reflective pixel electrodes for pixels of the device; a transparent layer bearing at least one transparent counter-electrode; and a layer of adjustable refractive index material between said counter-electrode and said pixel electrodes. The material has a first, isotropic refractive index, n, under no applied electric field and wherein under an applied electric field said material has a second, different refractive index, n′, wherein said second, different refractive index is isotropic perpendicular to a direction of said applied electric field.
According to an embodiment of the present invention the material is a non-liquid crystal material. An example of such a material is nitrobenzene.
According to an embodiment of the present invention the material is a blue-phase liquid crystal material and said backplane is an LCOS (liquid crystal on silicon) backplane.
The liquid crystal material may have an optical Kerr constant of greater than 5 nm/V2.
According to an embodiment of the present invention, the device comprises means to apply an ac bias voltage to one or both of said counter electrode and a set of said pixel electrodes. In an embodiment the ac bias voltage is at least 20V peak-to-peak.
In an embodiment the LCOS backplane incorporates a temperature sensitive element, the device further comprising a temperature control system coupled to said temperature sensing element to control the temperature of said material.
According to an embodiment of the present invention, the counter electrode is segmented into rows.
According to a third aspect of the present invention, there is provided a method of routing an optical signal. The method comprises providing a routing device comprising a layer of blue phase liquid crystal between a plurality of pixel electrodes and at least one counter electrode; applying a first spatial pattern of voltages between the pixel electrodes and the counter electrode such that the refractive index of the layer of blue phase liquid crystal varies spatially to provide a first diffraction grating that deflects the optical signal to a first output; and applying a second spatial pattern of voltages between the pixel electrodes and the counter electrode such that the refractive index of the layer of blue phase liquid crystal varies spatially to provide a second diffraction grating that deflects the optical signal to a second output, thereby switching the optical signal from the first output to the second output.
Embodiments of the invention allow polarisation independent routing with devices as described above.
In an embodiment, the method comprises applying a square wave alternating voltage to the at least one counter electrode; and applying a spatial pattern of drive voltages to the pixel electrodes, wherein the spatial pattern of drive voltages has a transition from a first set of drive voltages to a second set of drive voltages which coincides with a transition of the square wave alternating voltage and an instantaneous level of the square wave alternating voltage, at a first time, and said first set of drive voltages provide the first spatial pattern of voltages and an instantaneous level of the square wave alternating voltage, at a second time, and said second set of drive voltages provide the second spatial pattern of voltages.
In an embodiment applying a first spatial pattern of voltages between the pixel electrodes and the counter electrode comprises applying a first drive signal to the pixel electrodes and applying a second spatial pattern of voltages between the pixel electrodes and the counter electrode comprises applying a second drive signal to the pixel electrodes.
In an embodiment, the method further comprises applying an offset signal to the counter electrode. In an embodiment, the first drive signal and the second drive signals comprise a variable amplitude square wave.
In an embodiment, the method, further comprises applying a square wave alternating voltage to the counter electrode.
It has been observed that the refractive index certain of variable refractive index materials depends on the square of the voltage. By applying a voltage to both the counter electrode and the pixel electrodes and by timing the transitions of the voltages on the counter electrode with the transitions from the first drive signal to the second drive signal, the voltage applied across the variable refractive index material can be increased and the change in refractive index for a given change in pixel voltage can be maximised.
According to a fourth aspect of the present invention, there is provided a method of polarisation-insensitive switching or routing of one or more optical telecommunications signal beams. The method comprises: providing a reflective liquid crystal on silicon backplane having a plurality of pixels with a layer of blue-phase liquid crystal material over said backplane and a counter-electrode over said blue-phase liquid crystal material; and displaying at least one diffractive optical element by applying a patterned electric field to said blue-phase liquid crystal material, wherein said electric field runs between said pixels of said backplane and said counter electrode and substantially parallel to light incident onto and reflected from said reflective LCOS backplane; and deflecting a said signal beam using said refractive optical element.
In an embodiment the diffractive optical element is configured to deflect a said signal beam in two dimensions.
In an embodiment said diffractive optical element comprises a hologram, the method comprising switching or routing a plurality of said signal beams simultaneously using said hologram.
In an embodiment the signal beam comprises a wavelength division multiplex (WDM) signal beam, the method further comprising de-multiplexing said WDM signal beam into a plurality of separate wavelength signal beams, wherein said de-multiplexing retains components of two orthogonal polarising components in a said separate wavelength signal beam; directing said separate wavelength signal beams to different spatial regions of said diffractive optical element; deflecting said separate wavelength beams; and re-multiplexing at least some of said deflected separate wavelength beams.
In an embodiment said modulating using said electric field comprises applying a voltage of less than 25V between a pixel electrode of said LCOS backplane and said counter-electrode.
According to a fifth aspect of the present invention, there is provided a method of manipulating an optical signal. The method comprises providing a device comprising a layer of blue phase liquid crystal material in the path of the optical signal. By applying dynamically varying spatial pattern of voltages to the layer of blue phase liquid crystal material, the refractive index of the layer is caused to vary spatially according to the applied pattern.
In an embodiment, the optical signal is incident on the layer of blue phase liquid crystal at an angle of less than 5 degrees to normal. The inventors of the present invention have demonstrated that at angles of less than 5 degrees to normal, polarisation independent manipulation can be achieved with a variation in polarisation of less than λ/8.
The voltages applied to the blue phase liquid crystal material may be greater than 50 volts and preferably greater than 100 volts.
In an embodiment the device comprises a plurality of electrodes disposed on one side of the layer of blue phase liquid crystal material the dynamically varying spatial pattern of voltages is applied to the plurality of electrodes.
In an embodiment the device comprises a photoconductive layer and applying the dynamically varying spatial pattern of voltages comprises applying a light beam having a dynamically varying spatial pattern of intensities to the photoconductive layer such that the resistance across the photoconductive layer varies spatially .
Embodiments of the present invention employ optical addressing of the blue phase liquid crystal material layer.
In an embodiment, the method comprises measuring the optical signal to determine a disturbance in a wavefront of the optical signal and determining the dynamically varying pattern to reduce the disturbance. Such an embodiment provides polarisation independent adaptive optics.
According to a sixth aspect of the present invention, there is provided an optical element. The optical element comprises a first transparent electrode; a second transparent electrode; a variable refractive index layer disposed between the first and second transparent electrodes; and a photoconductive layer disposed between the second transparent electrode and the variable refractive index layer. The variable refractive index layer comprises a material having a first, isotropic refractive index under no applied electric field and a second, different refractive index under an applied electric field wherein the second, different refractive index is isotropic perpendicular to the direction of the applied electric field.
In an embodiment the variable refractive index layer comprises a blue phase liquid crystal material.
In an embodiment the element further comprises a light blocking layer between the variable refractive index layer and the photoconductive layer.
According to a seventh aspect of the present invention there is provided an adaptive optics system comprising an optical element as described above.
Features of the above-described aspects and embodiments of the invention may be combined in any permutation.
Anti-Reflection Structures/Coatings
In some preferred embodiments of each of the above described aspects of the invention the (routing) device or optical element incorporates one or more layers with an anti-reflection (AR) coating, for example on an upper transparent (optical input/output) layer of the device/element. Additionally or alternatively one or more internal interfaces of the device/element may also incorporate an anti-reflection coating. In some particular implementations, the characteristics of the AR coating vary over a surface of the device, in particular to optimise different spatial regions of the device/element for different operating wavelengths of the device/element. This is particularly advantageous in an optical system incorporating a wavelength-selective optical demultiplexer, to direct different optical wavelengths to the different spatial regions of the device/element optimised for those wavelengths.
In addition a diffraction pattern displayed on the device may incorporate a component to generate a reflection-cancelling beam, to partially or substantially wholly cancel a reflection from either an internal or external surface interface of the device/element. Preferably the diffraction pattern is configured so that the reflection-cancelling beam is generated by a second (and/or higher) diffraction order (the first diffraction order being used for the intended optical wavefront manipulation, for example routing).
The skilled person will appreciate that these latter techniques may be employed independently of the above described aspects of the invention.
Thus in a further aspect the invention provides an LCOS (liquid crystal on silicon) device comprising a surface bearing an anti-reflection structure, wherein: i) the anti-reflection structure comprises a physical surface having a topography with features having lateral dimensions of less than 2000 nm and having an average refraction index which decreases with distance away from said surface; and ii) a configuration of said topography, averaged over lateral dimensions of greater than 2000 nm, varies with lateral position on said surface.
In embodiments the surface topography comprises a pattern of features, in particular a regular/cyclical pattern of varying height or relief over the surface. Laterally the pattern repeats or has a characteristic length scale (in the case of a random pattern) which is less than the wavelength at which the device operates, for example less than 2000 nm, 1900 nm, 1800 nm, 1700 nm or 1600 nm. However the anti-reflection layer also has one or more parameters of the configuration or structure (for example, average feature pitch and/or feature shape) which, when averaged over distances greater than a wavelength, varies with position on the surface: that is there is a macroscopic variation of the topography. This allows different surface regions of the device to be optimised for different wavelengths, which is useful in many LCOS applications including, but not limited to: optical switching/routing/pulse shaping, holography for example for displaying images, the display of diffraction patterns/holograms for optical tweezers, and more generally any application where wavefront control is desired.
In preferred implementations the surface bearing the topography is a front surface of the device, more particularly a coverplate of the device, although the skilled person will recognise that this is not essential.
In a related aspect the invention provides an LCOS device, in particular as claimed in claim 34 or 35, combined with a controller to display a hologram on said LCOS device to deflect first light into a first diffractive order of said hologram, wherein said hologram is further configured to deflect second light into a second diffraction order of said hologram, wherein said LCOS device has an interface generating unwanted reflected light, and wherein said second light is in antiphase with said unwanted reflected light.
The features of such an LCOS device may be combined with or employed separately to the laterally varying surface topography pattern mentioned previously. Broadly speaking in embodiments, because the first diffractive order is employed for directing the light in whatever way is desired (for example, according to the application as described previously), a second and/or higher diffraction order may be employed to cancel one or more unwanted beams reflected from an internal interface. Thus in embodiments at least a proportion of the second light, and preferably substantially all the second light, is in antiphase with the reflected beam to be cancelled. Preferably the second (reflection-cancellation) beam has substantially the same power as the unwanted reflected light in that direction (the same amplitude and/or intensity). The skilled person will recognise, however, that some benefit may be obtained from the technique even where the beams are not exactly at the same power or exactly in antiphase.
The skilled person will appreciate that because holograms are additive in nature it is straightforward to implement the reflection cancellation beam: to a desired hologram displayed on the SLM it is simply necessary to add a second hologram (adding the respective pixel values) which generates a second (or higher) order beam with the desired power (amplitude/intensity), phase, and direction. The skilled person will be aware of many algorithms which may be employed to perform such a calculation. The power, phase and direction of the unwanted reflection to be cancelled can be readily obtained by numerical modelling of the LCOS device using any of a number standard optical modelling packages—in general the precise amplitude, phase and direction of a reflection to be cancelled will depend upon the detail structure of the LCOS device.
As previously mentioned either or both of the above approaches may be employed to optimise the anti-reflection properties at different respective wavelengths in different respective lateral surface regions of the device.
In a related aspect the invention provides a method of suppressing an unwanted reflection in a spatial light modulator (SLM) comprising a reflective liquid crystal on a silicon backplane having a plurality of pixels with a layer of blue-phase liquid crystal material over said backplane and a counter-electrode over said blue-phase liquid crystal material, the method comprising using said SLM to deflect light into a second or higher diffraction order of a diffraction pattern displayed on said grating, wherein said second or higher order deflected light defines a reflection-cancellation beam in antiphase with said unwanted reflection.
Preferably, as previously described, the SLM is an LCOS SLM. Preferably the unwanted reflection comprises an internal reflection at an interface of the device, although in principle the technique is not restricted to such an internal reflection. The interface may be an internal interface, for example a coverplate/liquid crystal interface such as an interface comprising one or more of a liquid crystal layer, an alignment layer, an electrode layer, and a coverplate layer. The skilled person will appreciate that, depending upon the configuration of the device and/or the thickness of the various layers, the interface may or may not include all of these. For example, in a blue phase device the alignment layer is not required. The reflection cancellation need not be complete, although preferably it is substantially complete.
In a still further related aspect the invention provides a spatial light modulator (SLM) in combination with an SLM controller, said SLM having an interface generating an unwanted reflection, wherein said SLM controller is configured to drive said SLM to deflect light into a second or higher diffraction order of a diffraction pattern displayed on said grating, wherein said second or higher order deflected light defines a reflection-cancellation beam in antiphase with said unwanted reflection.
In the following embodiments of the invention will be described, by way of example only, with reference to the accompanying drawings in which:
The adjustable refractive index material 106 has a refractive index which varies with the electric field applied across it. An example of such a material is blue phase liquid crystal. The adjustable refractive index material 106 has an isotropic refractive index in the direction perpendicular to the applied electric field.
In the device shown in
By switching between different patterns of voltages as described above, an incident beam of light can be switched between different outputs.
If a material that has an isotropic refractive index in the direction perpendicular to the applied electric field is used, the switching is independent of the polarisation of the incident beam.
In the following, the wording isotropic perpendicular to the applied electric field is used to describe that there is substantially no difference in refractive index in any direction perpendicular to the applied electric field, the refractive index may vary in any other direction.
Deforming the domains (e.g. by applying an electric field) along the propagation direction of the light can be used to modulate polarised light, since the deformed domain undergoes a change in optical refractive index and becomes optically anisotropic. If the deformation is along the light transmission axis so that the deformed domain is still symmetrical normal to the light transmission, a localised change in refractive index occurs which is independent of the incident polarisation angle, and can be used to produce SLMs capable of polarisation independent phase modulation.
The behaviour discussed above was experimentally demonstrated as follows. A liquid crystal LCM-PSBP.1328UV (from LC Matter Corp.) was used for the experiment. A 12 micron thick glass cell with ITO electrodes and no other alignment layers was filled with the mixture. Square waveform voltage with frequency of 1 kHz was applied across the cell in order to introduce anisotropy in the Blue Phase.
Phase shift was measured using a cell as a Fabry-Perot interferometer. The cell and detected intensity are shown in
The high voltage required to get 2π phase shift reflects the fact the material has low Kerr constant (K) of 1.6 nm/V2 at 600 nm. The drive voltage and cell thickness should be reduced considerably by using materials with the large Kerr constant (K up to 12.7 nm/V2) that have been recently developed for in-plane switching display applications. For this material a 2π phase shift at 600 nm wavelength should be obtained with a 10 μm reflective cell and driving voltage of ±14V, well within the range of LCOS backplanes.
Thus, the inventors have demonstrated a method for phase modulation of randomly polarised light.
It is envisaged that the liquid crystal material may be chosen in order to get a wide temperature range blue phase. Mixtures of a liquid crystal and a polymer have been shown to stabilise the blue phase in a temperature range from 60 C to −10 C degrees. In order to stabilise the blue phase in a temperature range, the blue phase material may be doped with nanoparticles. This improves the sensitivity of the blue phase material to electric fields.
Another possibility is the use of bimesogenic liquid crystals.
The Blue Phase has the additional advantage that no alignment or rubbing layers are required. This makes possible simplified polarisation-independent telecoms routing.
The second method, digital variable mark/space ratio drive, involves applying a square wave to the front electrode of the same amplitude as the supply voltage, and the pixels are switched at different times between the supply or ground. This allows a dc-balanced but asymmetrically timed waveform across the LC which has a variable RMS value, to which the LC responds, giving an average response similar to the fixed FE version, but with small variations as the LC follows the waveform.
The latter method is satisfactory for image projection applications, but causes unacceptable phase flicker which decreases the efficiency of phase holograms. However it allows twice the drive voltage across the liquid crystal from the same backplane process, or for the same voltage requires a lower voltage process, allowing smaller transistors and therefore higher integration, lower power, and lower cost.
Since rows of pixels are usually loaded sequentially down the pixel array a combination of the two causes progressive loss of drive and increased phase flicker down the pixel array.
The blue phase liquid crystal responds more rapidly to the applied electric field than a nematic phase liquid crystal. As shown in
As can be seen in
However for this to work on a large 2-dimensional pixel array, all the rows of pixels have to be loaded on the FE transition, otherwise phase flicker will be reintroduced which will get progressively worse down the rows of refreshed pixels.
The chip can be designed so that rows of rectangular pixels forming 1-dimensional arrays, suitable for grating-type holograms, can be driven, with a horizontally segmented FE. This allows a separate driver for each FE segment, ensuring that the FE switches at the same time that the pixels are refreshed, eliminating the phase flicker, and allowing the LC drive to be offset up the response curve.
For a multiple wavelength routing device a system can be designed to ensure that the separated wavelengths fall onto the required segments. Such a system requires a reasonably polarisation independent wavelength splitting grating and allows a single-device polarisation independent router.
The router 800 switches signals between fibres of a fibre array 802. An input signal is received in an input fibre 802. This is switched to one or more output fibres depending on the state of the switch. The input signal is passes through a lens 806 and is wavelength demultiplexed by wavelength demultiplexer 808. Different wavelengths are directed to different areas of the LCOS device 812. The LCOS device 812 applies a hologram/grating which deflects the wavelength incident on that particular area in the direction orthogonal to the wavelength splitting axis. The beam is reflected through a wavelength multiplexer 814 which collects together all the wavelengths which have been deflected by a given angle and these are then focussed by a lens 816 into the corresponding output fibre 818. The angle of deflection (inversely proportional to grating pitch) determines the destination fibre.
This LCOS device can be segmented by having strips of ITO on the cover glass each corresponding to a separate wavelength channel, so that each wavelength can be driven independently of other wavelengths.
Beneath each front electrode strip there is a one-dimensional strip having the same width as the front electrode strip. The pixels may be rectangular as shown in
The segments which form the diffraction grating can be up to 20 mm high and unto 100 microns wide.
The front electrode connections are made by silver loaded epoxy from the ITO front electrodes on the front glass cover.
A read beam 1240 is incident on the substrate 1202 on the side of the device including the adjustable refractive index layer 1206. The variation in the voltage across the variable refractive index layer 1206 causes the refractive index of the variable refractive index layer to vary spatially. Therefore, the phase of the outgoing light 1242 reflected from the reflective layer 1210 can be modified. The reflective layer 1210 stops the write beam from passing through the device and the read beam from activating the photoconducting layer.
The voltage waveform 1220 may take the form described in reference to
In place of or in addition to the reflective layer, the device may also comprise a light blocking layer. Such a light blocking layer could be arranged below the reflective layer and configured to absorb the 1-5% of residual light and stop it hitting the photoconductive layer.
In an alternative embodiment, the device may be used in a transmissive configuration. In this configuration, there is no reflective layer and the read beam of light passes through the device. In such an embodiment, different wavelengths of light are used for the read and write beams of light or each beam is pulsed in a different part of the voltage waveform.
The device described above has applications in adaptive optics. Light travelling through a medium can be disrupted by variations in refractive index—e.g. in the atmosphere due to turbulence arising from pressure changes or convection caused by temperature changes. For any object viewed in the far field, this alters the shape of the plane wave across the entry optics, and limits the resolution of the optics below the theoretical limit. E.g. for an astronomical telescope this gives a blurred jittering image of a star which should be a point source. For planetary or surveillance images this also gives instantaneous spatial distortion.
Adaptive optics can partially compensate for the disturbance, usually by using a deformable mirror with an array of electro-mechanical actuators. These can correct the wavefront deformation of a few microns and partially restore the wavefront.
The input waveform 1508 is partially split by a beamsplitter 1510. Part of the beam is incident upon a wavefront sensor 1512 and the remaining part of the beam is incident upon an active phase correction device 1514 which comprises a layer of variable refractive index material such as blue phase liquid crystal as described above. The output of the wavefront sensor 1512 is used by an adaptive optics correction controller 1516 to control the spatial variation of the voltage applied across the variable refractive index material and thereby control the phase correction. The corrected beam 1518 is output to imaging optics of the telescope.
Adaptive optics can also be used to correct for aberrations in optical systems , improving vision to a greater extent than lens correction, and e.g. for laser surgery, when the imperfections of the eye's lens must be corrected for retinal welding, etc.
Many adaptive optics applications require polarisation independence and a rapid (millisecond) response time, the devices described above are suited to these applications since they allow for polarisation independent phase correction without a requirement for moving parts such as adjustable mirrors.
The devices described above provide polarisation independent phase modulation for light beams of normal incidence. For beams with a non-zero angle of incidence, some polarisation dependence is introduced into the phase change.
As can be seen from
Anti-Reflection Structures/Coatings
We will now describe how the performance of blue phase and other LCOS SLMs can be enhanced using structured surfaces on the front SLM cover-plate to reduce reflections. This is particularly applicable to wavelength selective switches, where a very low reflectivity over a large bandwidth is desirable.
The example LCOS SLM of
For most applications, the front surface reflectivity can be sufficiently reduced through the use of standard thin film coatings. However, for telecom applications, such as the implementation of wavelength selective switches based on beam-steering, the reflectivity of the front surface should preferably be reduced to Rf<0.01% over a wide wavelength range in order to minimize crosstalk. To illustrate this consider
where λ is the wavelength of the light and the angles are the values measured in air. For an ideal blazed grating, all the light is diffracted into the m=+1 order. However, due to spatial and phase quantization of the grating, some light ends up in higher and symmetric (m≠1) orders. For certain switch geometries, light in these m≠1 orders can couple into output ports leading to crosstalk. For practical telecom applications the crosstalk should preferably be suppressed to <−40 dB.
Consider the case where θi=0. As mentioned, ideally all the light is diffracted into the +1 order through an angle of θ1. However, if the front surface of the SLM has a finite reflectivity of Rf, a portion of the +1 order will be reflected back towards the SLM at an angle −θ1 with respect to the normal. This order will be re-diffracted by the grating such that it propagates at an angle of 2θ1 to the normal, which, for small diffraction angles, corresponds to the angle of propagation of the m=+2 order of the original diffracted beam, θ2. Assuming a coverplate with refractive index of n=1.5, and no anti-refection coating, the Fresnel reflection coefficient of the front coverplate surface is approximately 4%. Let us assume an SLM which has 100% diffraction efficiency (no absorption losses and a perfect blazed grating). The power in the m=+1 and m=+2 directions will therefore be given by 0.96Pin and 0.04Pin respectively (ignoring multiple reflections). As a result, the theoretical crosstalk is −13.8 dB. This compares well to values of −14 to −18 dB measured experimentally. If we reduce the front face reflectivity to 1% (typical of single layer thin film coatings), the crosstalk reduces to approximately −20 dB. To reduce the crosstalk to <−40 dB the front face reflectivity should preferably be reduced to <0.01%. Such a low reflectivity is challenging, particularly as we should preferably ensure this value is maintained across the C-band, L-band, or C and L band. For example, the C-band extends from 1530 nm to 1570 nm, and L-band from 1565 nm to 1625 nm, and an optimized multi-layer coating can expensive to fabricate.
Although particularly important for telecom applications, other applications that employ the optimization of complex hologram patterns also benefit from a reduction of front face reflectivity. The quality of the replay field in display holography and optical tweezers will improve as reducing Rf ensures that the output wavefront more closely matches the designed wavefront.
It is desirable to provide an anti-reflection layer with the following properties:
There are four potential techniques which may be employed for reducing crosstalk in practical LCOS SLM applications: geometric, thin film dielectric coatings, graded index coatings, and patterned nano-structures:
Graded Index/Patterned Nano-Structures for Telecoms Wavelengths
Raguin and Morris showed theoretically that a multi-level approximation to a pyramidal anti-reflection structure fabricated on a GaAs substrate (N=2, 4, and 8 phase-levels) and optimized for operation at 10.6 μm could have a reflectivity <3×10−3% for all values of N for a randomly polarized normally incident beam (D. H. Raguin and G. M. Morris, “Antireflection structured surfaces for the intrared spectral region”, Appl. Opt. 32, 1154-1166 (1993)). This exceeds the above target of 0.01%. Moreover, their model predicted a T=99.9% transmission when N=8 even if the wavelength range varied by 10.6 μm±10%, and the incident half angle varied by 30°.
Experimental results for graded index coating and patterned nano-structures have been presented for a variety of surface topologies, material systems, wavelengths and operating conditions (see the references in P. Lalanne and M. Hutley, “The optical properties of artificial media structured at a subwavelength scale”). Surface fabrication and replication techniques allow such surfaces to be fabricated for the visible. For example, Hutley and Gombert have used UV embossing into plastic using a nickel master to produce AR coated Fresnel lenses for overhead projectors (M. Hutley and A. Gombert, “Moth-eyes: the tortuous path from a glint in the eye to a commercial reality”, Photonics Science News 6, 35-39 (2000)). Anti-reflection nano-structures may be produced for the optical C- and L-bands by similar techniques.
Referring again to
Example Nanostructured Surfaces
Broadly speaking, the aim of using a nano-structured anti-reflection coating is to make a more effective broadband coating for telecoms, and one that can be tailored so that the reflectivity is <0.01% at each wavelength location across the SLM in a system that uses a wavelength de-multiplexer. This will allow simplification of the hologram design and optimization by reducing the front surface reflections (so they may be neglected), particularly coupled with the fact that blue phase SLMs do not require an alignment layer, thereby simplifying calculation and measurement of the coverplate/electrode /liquid crystal interface reflectivity.
For details of how to implement a wavelength-optimised coating reference may be made to A. Deinega, I. Valuev, B. Potapkin, and Y. Lozovik, “Minimizing light reflection from dielectric textured surfaces”, J. Opt. Soc. Am. A, Vol. 28, No. 5, pp. 770-776, May 2011. In this paper simulations of the reflectivity from a variety of nano-structured surfaces are modelled using finite difference time domain analysis (FDTD). This technique gives a numerical solution to Maxwell's equations for complex surfaces by probing the structure with a plane wave in the form of a short pulse of light. As the pulse has a certain wavelength spread, on analysis of the reflected field the reflectivity as a function of wavelength can be derived. They model pyramidal surface relief structures which have a depth of d, a width of 2 L and a period of A over a full range of wavelengths on a glass substrate (n=1.5). We are interested in the long wavelength limit (Λ<λ) as this minimizes scattering losses as the light sees an effective graded index interface. For Λ˜λ and Λ>λ we would get scattering and diffraction losses as the light would interact with effectively a periodic macro-prism structure on the surface of the SLM. This regime is considered in the paper, and in particular Deinega et al analyse square pyramids with linear, cubic and quantic profiles where Λ=2 L. The two key results from this paper are in
As mentioned we are interested in the long wavelength region where λ>Λ in order to minimize scattering. There are two ways to use this graph. Firstly for a varying nano-structure depth, d, and secondly for a fixed nano-structure depth. Let us assume that we have two discrete wavelengths, λ1 and λ2.
As blue phase liquid crystal based SLMs do not need alignment layers, this simplifies the suppression of crosstalk from the coverplate/ITO/liquid crystal surface as it is easier to calculate the theoretical reflectivity of this surface and, optionally, incorporate this in the hologram design.
Coverplate/Liquid Crystal Interface
We now consider the reflectivity due to the coverplate/liquid crystal (comprising glass, electrode (for example ITO—indium tin oxide), optional alignment layer, and liquid crystal layer(s)). This is more complicated because potentially four interfaces are present, and the alignment layer/liquid crystal layer reflectivity depends on the state of the liquid crystal. However, it is possible to ‘tune out’ the coverplate/liquid crystal reflectivity by employing a hologram pattern which sends an equal amount of power into the +2 order 180° out-of-phase with the power due to the coverplate/liquid crystal reflectivity. As the liquid crystal layer is very thin, temperature changes have negligible effect on the path length, so the reflection-cancellation hologram needs only be determined once. This approach is possible but not so easily implemented when dealing the coverplate front surface reflectivity, which is −1 mm thick, as the same temperature change induces a larger optical path length change. To illustrate this, the optical path length change, Δopl, is given by Δopl=ndαΔT, where n is the index, d is the material thickness, α is the coefficient of thermal expansion, and ΔT is the temperature change. Thus Δopl is proportional to ΔT, and maintaining the 180° out-of-phase condition becomes progressively harder as d increases.
The reflection-compensation, HR, hologram is computer generated, for example by any of a range of standard techniques (for example a Gerchberg-Saxton algorithm) and stored in non-volatile memory. This is then added to the target hologram HT to provide a combined hologram HC=HR+HT for display.
One way to compensate for the coverplate/ITO/alignment layer/liquid crystal interface reflectivity is, if the fundamental blazed grating has a period of T, to add a weak blazed grating (a compensating grating) of period 2T. This generates a +1 order beam that travels in the same direction as the +2 diffracted beam of the fundamental blazed grating (which also includes the unwanted reflected light as shown in
As previously mentioned, applying this cancellation technique to the front coverplate surface is more difficult due to the relative thickness of the coverplate. Temperature changes can cause the phase of the reflected beam to change with respect to the cancellation beam. However this may be compensated for by correcting the hologram for temperature, using a very thin coverplate, and/or by using a low thermal expansion coefficient glass.
We have described, in embodiments, a phase-only LCOS device employing liquid crystal in a blue phase. It will be understood that the invention is not limited to the described embodiments and encompasses modifications apparent to those skilled in the art lying within the spirit and scope of the claims appended hereto.
| Number | Date | Country | Kind |
|---|---|---|---|
| 1201190.4 | Jan 2012 | GB | national |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 14374628 | Jul 2014 | US |
| Child | 15435181 | US |
