WO2021113898A1

WO2021113898A1 - Electromagnetic filter device and method of use

Info

Publication number: WO2021113898A1
Application number: PCT/AU2020/051245
Authority: WO
Inventors: Lukas WESEMANN; Ann ROBERTS; Timothy John Davis
Original assignee: The University Of Melbourne
Priority date: 2019-12-10
Filing date: 2020-11-18
Publication date: 2021-06-17

Abstract

An electromagnetic filter device comprising a substrate and a high refractive index dielectric layer (HRI layer) located on a top surface of the substrate, and an arrangement of patch elements extending past a top surface of the HRI layer, wherein the patch elements are arranged to cause a coupling of incident electromagnetic radiation to the HRI layer, and wherein the HRI layer is configured to directly couple electromagnetic radiation at each patch element to one or more other patch elements.

Description

ELECTROMAGNETIC FILTER DEVICE AND METHOD OF USE

Field of the Invention

The invention generally relates to an electromagnetic filter device.

Background to the Invention

Cameras generate images which are used in applications as diverse as optical microscopy for industrial and medical purposes, self-driving cars and remote sensing for agriculture and security using unmanned aerial vehicles (drones) or satellites. Images not only form part of the acquired data but are also critical to the control of the system itself.

Raw images, however, can contain noise or data extraneous to the relevant application or important information is effectively ‘invisible’. For this reason, digital and optical image processing techniques have become ubiquitous and algorithms and optical methods for noise reduction, edge detection, image enhancement and object identification are widely used.

Digital approaches, however, cannot access the phase of a light field and also require energy and time which scale with the extent of the data acquired. While optical solutions to visualize the phase of a wavefield exist, most commonly used for biological imaging, they comprise bulky and costly components. For this reason, conventional optical methods cannot be integrated in compact devices and platforms such as next generation mobile medical equipment, smartphones and nanosatellites.

Summary of the Invention

According to an aspect of the present invention, there is provided an electromagnetic filter device comprising a substrate and a high refractive index dielectric layer (HRI layer) located on a top surface of the substrate, and an arrangement of patch elements extending past a top surface of the HRI layer, wherein the patch elements are arranged to cause a coupling of incident electromagnetic radiation to the HRI layer, and wherein the HRI layer is configured to direct coupled electromagnetic radiation at each patch element to one or more other patch elements. Optionally, the arrangement of patch elements is configured to implement a corresponding optical transfer function with respect to light incident at a preselected angle. The arrangement of patch elements may be configured to implement a high-pass spatial frequency filter.

Optionally, the patch elements are arranged with a constant period in at least one direction, wherein the period is selected based on at least an intended wavelength of the electromagnetic radiation. The patch period may be smaller than the wavelength of the intended electromagnetic radiation.

In an embodiment, the HRI layer comprises Ti02. Also, in an embodiment, the patch elements comprise a metal or a dielectric. The patch elements may be formed on top of the HRI layer. The patch elements are arranged in a regular array, such as a two-dimensional rectangular array. The regular array may be a one-dimensional array comprising elongate patch elements.

Optionally, the device comprises a cover layer applied to the top layer of the HRI layer such as to cover the patch elements, wherein the refractive index of the cover layer is lower than the refractive index of the HRI layer. The cover layer may comprise polymethyl methacrylate (PMMA), a glass, or SiC .

In an embodiment, the arrangement of patch elements and HRI layer are configured that the electromagnetic filter device shows increased transmission with respect to one angle of incidence when compared to another angle of incidence of the incident electromagnetic radiation.

The arrangement of patch elements may be configured to interact with a first polarisation of incident light differently to a second polarisation of light. The device may be configured to show a higher transmission of the first polarisation when compared to the second polarisation, and the first polarisation may be orthogonal to the second polarisation.

The device may be configured for use with a biological sample

The substrate may comprise a glass, S1O2, or polymethyl methacrylate (PMMA). The device may be integrated with a glass microscope slide. Alternatively, the device may be adapted for use as an optical component of a camera.

According to another aspect of the present invention, there is provided a method of imaging a sample using an electromagnetic filter device according to the previous aspect, comprising positioning the sample between an electromagnetic radiation source and the electromagnetic filter device, illuminating the sample with electromagnetic radiation from the electromagnetic radiation source, and capturing an image of the electromagnetic radiation transmitted through the electromagnetic filter device. The sample may be a biological sample.

According to another aspect of the present invention, there is provided a filter attachment for a camera, microscope, or optical system, wherein the filter attachment comprises the electromagnetic filter device of the first aspect.

According to another aspect of the present invention, there is provided a camera, microscope, or optical system comprising the electromagnetic filter device of the first aspect.

As used herein, the word “comprise” or variations such as “comprises” or “comprising” is used in an inclusive sense, i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.

Brief Description of the Drawings

In order that the invention may be more clearly understood, embodiments will now be described, by way of example, with reference to the accompanying drawing, in which:

Figure 1 shows an electromagnetic filter device according to an embodiment;

Figure 2 shows an example of an arrangement of patch elements;

Figure 3 shows an example arrangement of electromagnetic filter element, electromagnetic source, and sample;

Figure 4 shows a schematic representation of coupling of incident scattered electromagnetic radiation;

Figure 5 shows an example of simulated transmittances for various combinations of patch element period and incident wavelength;

Figure 6 shows a calculated plot of transmitted amplitude vs angle of incidence for different patch element heights;

Figure 7 shows simulated results for different duty cycles for a high-pass filter device;

Figure 8 shows a simulated demonstration of the edge enhancement provided by the high- pass filter device; Figure 9 shows an experimentally generated image due to a high-pass filter device;

Figure 10 shows an example of a device configured to convert phase gradients into an intensity modulation;

Figure 11 shows representative SEM images of manufactured devices;

Figure 12 shows reflectance of a device operating in reflection mode

Figure 13 shows patch elements having a ring geometry;

Figure 14 shows a method for manufacturing a device according to an embodiment;

Figure 15 shows an example of a one-dimensional arrangement of patch elements;

Figure 16 shows transmission of a Fourier plane wave through the device of Figure 15;

Figure 17 shows imagining of a sample using the device of Figure 15;

Figure 18 shows an effect of tilting an electromagnetic filter device with respect to incident light;

Figure 19 shows imaging of a sample comprising HeLa cells; and

Figure 20 shows an experimental result for differently configured devices 10 illuminated by plane-waves.

Description of Embodiments

Figure 1 shows an electromagnetic filter device 10 according to an embodiment. The device 10 comprises a substrate layer (substrate) 11 and a high refractive index dielectric layer (HRI layer) 12 located on a top surface of the substrate 11. The HRI layer 12 itself includes a top surface located distally to the substrate 11. An arrangement of patch elements 13 extends above the top surface of the HRI layer 12. In the embodiment shown, the patch elements 13 are formed directly onto the HRI layer 12. In another embodiment, the patch elements 13 are formed from or directly on the substrate 11 and extend through the HRI layer 12. Relevantly, the refractive index of the HRI layer 12 is higher than the refractive index of the substrate 11. The substrate 11 can be any suitable material, for example the substrate may comprise glass, silicon dioxide (SiCh), or polymethyl methacrylate (PMMA).

Reference to “top surface” herein is intended to clarify a relative relationship of the surface of the relevant layer to other layers — the top layers are those indicated in the direction “TOP” in Figure 1. General features shown in the figures are referred to with a numerical reference — equivalent features in different drawings are labelled with the same value. Where necessary to distinguish particular instances of a general feature, a lowercase letter suffix is used. For example, in Figure 1, reference is made to the general feature of a patch element 13 while specific reference can be made to patch elements 13a, 13b, 13c, 13d.

Figure 1 also shows an optional cover layer 14 applied onto the patch elements 13 and the top surface of the HRI layer 12. The cover layer 14 can provide a protective function to the underlaying device features. The cover layer 14 typically will have a refractive index lower than the HRI layer 12 — for example, the refractive index may be substantially the same as the substrate 11. The cover layer 14 can comprise PMMA, which may be 0.75 - 1 pm thick.

The patch elements 13 are configured to interact with incident electromagnetic radiation and, to this end, may comprise a metal (for example Aluminium) and/or a dielectric material. In a variation, the HRI layer 12 includes a surface relief corresponding to the patch elements 13. In a variation (not shown), the patch elements 13 are formed from recessions in the HRI layer 12 — these recessions may comprise a metal or dielectric, thereby enabling interaction with the incident light.

Referring to Figure 2, the arrangement of patch elements 13 can define a regular array, for example, a rectangular array. In the embodiment shown, the patch elements 13 are arranged as a rectangular array with constant period (P) in both perpendicular directions X and Y. It should be noted that the figure shows a plan view looking towards the top surface of the device 10. The patch elements 13 have an element length (L) and are arranged with patch distance (D) between adjacent elements 13. Therefore, the period is equal to the sum of the patch distance and element length:

P = L + D

Eq. 1

The duty cycle (C) can be defined as the ratio of patch period to element length:

C = ^L/p

Eq. 2

The aforementioned parameters P, L, D, and C can be selected according to a desired characteristic of the electromagnetic filter device 10, as is discussed below. For the purpose of this disclosure, the rectangular arrangement shown in Figure 2 is characterised by its duty cycle C and period P.

In another embodiment, the duty cycle C and/or period P may be varied over the surface of the device 10. In another embodiment, the duty cycle C and/or period P may be different in different directions. These embodiments may facilitate spatial-frequency-dependent filtering or different filtering of optical signals. It is also expected that non-rectangularly symmetric arrangements may be employed, for example, a circularly symmetric arrangement of patch elements 13 or a hexagonal arrangement of patch elements 13. In an embodiment, the period is lower than the free-space wavelength of the electromagnetic radiation.

Figure 3 shows two arrangements, one corresponding to a device 10 configured to operate in a transmission mode and the other corresponding to a device configured to operate in a reflection mode. An electromagnetic source 50 typically provides a monochromatic source of electromagnetic radiation — for convenience of description, unless otherwise noted, the term “light source 50” refers to the electromagnetic source 50. The light source 50 may generate electromagnetic radiation within the visible spectrum or outside the visible spectrum.

The device 10 is configured to receive electromagnetic radiation which has undergone an interaction with the sample 51 — in the embodiment shown, the sample 51 is located between the light source 50 and the device 10. The device 10 can be placed at other points in the beam path. In the transmission mode, an imaging sensor 52 is provided on an opposite side of the device 10 to the light source 50 and sample 51. In the reflection mode, the device 10 includes a reflective layer on an opposite side to that receiving illumination from the light source 50, thereby causing a reflection of light received by the device 10. This reflected light can then be imaged by an imaging sensor 52. In an implementation, the sample 51 is placed onto the top surface of the device 10 — this surface may correspond to the cover layer 14 or another layer applied to the cover layer 14. It should be noted that, in certain embodiments, the device 10 may be reversed (i.e. so that its bottom side is facing the light source 50) while still providing the functionality herein described.

The sample 51 can interact with the incident light in one or more ways, depending on the nature of the sample 51. For example, a coherent incident planar light field may be modified by blocking portions of the incident light and/or modify the amplitude and/or modifying the phase of the incident light. The effect of the sample 51 is therefore to introduce changes to the spatial frequencies of the incident light field — here, the “spatial frequencies” corresponds to the spatial frequencies with respect to the plane of the device 10. Spatial frequencies correspond to the decomposition of the light field transmitted through the sample 51 into sine or cosine waves with different periods, where a low spatial frequency corresponds to a large period and slowly varying features in the sample 51 and a high spatial frequency corresponds to a small period and fine features such as edges in the sample 51. The device 10 selectively blocks or transmits different spatial frequencies (introduced by the sample 51 interacting with the incident light field) such that the light transmitted through the device 10 creates a filtered image of the sample 51. That is, the projection of the sample 51 onto the top of the device 10. The spatial frequency may be termed “lateral spatial frequency” or “transverse spatial frequency” as it is the spatial frequency extending laterally or transversely across the surface of the device 10. The projection can be related to a phase gradient across the surface of the device 10 of the incident light field. That is, spatial frequency refers to the lateral (in plane, on the surface) components k_x and k_y of the wavevector [k_x, k_y, k_z] of the incident light with a total length of ko. The spatial frequency components k_x and k_y may directly depend on the angle of incidence of the light relative to the surface of the device.

Without intending to be bound by any particular theory, referring to Figure 4, it is believed that each patch element 13 acts to scatter at least a portion of the incident light into the HRI layer 12 via a scattering process. At a particular patch element 13, the proportion of scattered light may depend on a local phase gradient. The HRI layer 12 acts as a waveguide (e.g. a slab-waveguide), providing a mechanism for the scattered light to interact with light scattered by other patch elements 13. In order for the HRI layer 12 to provide the function of a slab-waveguide, it may be preferred that the refractive index of the HRI layer 12 is at least 5% higher than that of the substrate 11. In an embodiment, the HRI layer 12 corresponds to a T1O2 layer. It may also be that light scattered by a patch element 13 is coupled directly to an adjacent patch element 13 (i.e. not via the HRI layer 12) — this is also shown in green in Figure 4. The scattering may correspond to a diffraction effect.

The light scattered by a particular patch element 13 propagates within the HRI layer 13 and eventually interacts with other patch elements 13. Similarly, the particular patch element 13 will receive light scattered by the other patch elements 13. The light so received at a patch element 13 can scatter at that patch element 13, allowing it to exit the HRI layer 12 and, ultimately, the device 10. The exit scattering at a particular patch element 13 depends on the total phase of all the waves reaching it, both within the HRI layer 12 and from adjacent patch elements 13. In this way, information regarding the local phase gradient at one patch element 13 is communicated throughout the device 10.

The device 10 is designed according to an intended use of the device 10, which may comprise use under illumination by a particular wavelength of light. The device 10 can, in a general sense, be configured to apply a desired optical transfer function based on tuning of one or more of the parameters of the device 10. An optical transfer function can correspond to a desired characteristic image enhancement — for example, low- or high-pass filtering, bandpass filtering, or some other filtering operation. A suitable structure for providing the desired optical transfer function may be determined using known electromagnetic modelling techniques. For example, the high-pass filter described below was modelled using COMSOL Multiphysics 4.5a - a Finite Element Method simulation.

An application of the device 10 as a high-pass spatial frequency filter is now described. A high-pass spatial frequency filter has the property of transmitting higher spatial frequencies (i.e. those representing edges, for example) while suppressing transmission of lower spatial frequencies. This is equivalent to preferentially transmitting light of a particular wavelength incident at an angle to the normal (Q ¹ 0°) to that same light incident at the normal (Q = 0°).

Referring to Figure 5, the effect on different wavelengths of normally incident light (Q = 0°) is shown for different periods P (assuming a rectangularly symmetric arrangement of patch elements 13 with equal periods in the two directions). As can be seen, it is possible to select a period P for a particular wavelength to either maximise or minimise transmission (or of course, select for some intermediate relative transmission) at normal incidence. Similarly, it is possible to select a wavelength to maximise or minimise transmission for a particular period P. For example, for a particular wavelength of normally incident light, the grating period may be selected such that light scattered by a first patch element 13 and subsequently scattered by another patch element 13 exits the device 10 having an opposite phase to the incident light. In this way, the twice scattered light destructively interferes with non-scattered incident light, minimising the intensity of the light exiting the device 10 at the normal.

For this particular simulation, the fixed parameters include a duty cycle C of 0.6 and a patch height H of 40 nm, for a T1O2 HRI layer 12 having refractive index 2.25 and silver patch elements 13. The thickness of the HRI layer 12 was set at 100 nm. The points marked [1], [2], and [3] constitute points of minimum transmission for patch periods P of 350 nm, 400 nm, and 450 nm, respectively. These specific results are illustrated in Table One, below.

Table One: Simulation Results That is, suitable normally incident wavelengths for minimum transmissions were calculated for particular periods P. Alternatively, the opposite calculation may be made — determining a suitable period P for a particular wavelength l.

It is possible to include experimental data when designing a particular device 10 — that is, the simulations may be utilised in order to determine several candidate configurations, which can subsequently be tested for efficiency and performance.

The simulation of Figure 5 provides a starting basis for additional analysis — the values of duty cycle C and patch height H, in particular, may be configurable to improve results.

It is generally found that, with a particular arrangement of patch elements 13, electromagnetic radiation incident at an angle not equal to the normal (i.e. Q ¹ 0°) interacts with the patch elements 13 such that a different proportion of the amplitude is redirected through the

HRI layer 12 when compared to normally incident radiation. The proportion may increase and/or decrease with angle Q.

Referring to Figure 6, transmission of a device 10 for unpolarised incident electromagnetic radiation, the device 10 having grating period P = 400 nm and duty cycle C = 60% as a function of wavelength and patch element heights (H) is shown at (a) and a calculated plot of transmitted amplitude as a function of spatial frequency k_x/ko for various patch element heights (H) (also referred to as “patch thickness”) is shown at (b). As can be seen, from k_x/ko = 0 up to about k_x/ko = 1.07, the relationship between angle of incidence and normalised transmission is monotonic. The result is, for a suitably prepared device 10 operating under an illumination by a light source 50 with characteristic wavelength, the device 10 will act a high- pass spatial frequency filter due to the preference for transmitting light incident at an angle (as discussed previously, associated with higher spatial frequencies) — with higher transmission the further the incident light is from the normal.

Regarding design of the device 10, it can be seen that the highest contrast of the heights H simulated is for patch element heights of 40 nm. Other heights also provide a monotonic increase over the same range, but with reduced contrast.

Figure 7 shows simulated results for transmission of unpolarised incident electromagnetic radiation for different duty cycles C for the high-pass filter device 10, the device 10 having grating periods (P) of (a) 350 nm, (b) 400 nm, and (c) 450 nm. The simulations are for a T1O2 HRI layer 12 with thickness 100 nm. As can be seen, the highest contrast is for a duty cycle C of 0.5. In fact, several of the duty cycles C show clearly unwanted behaviour for a high pass filter.

Figure 8 shows a simulated demonstration of the edge enhancement provided by the high- pass filter device 10. The sample 51 shown is of a depiction of the number “2” (labelled “input”). The three horizontal rows are for different values of period P and wavelength, based on the simulated results summarised in Table One (i.e., (a)-(c) P=350 nm and l=587 nm, (d)- (f) P=400 nm and l=645 nm, (g)-(i) P=450 nm and l=705 nm). The three vertical rows are for different polarisations of incident light, as labelled. The arrangement is for incident light, normal to the device 10, interacting with the sample 51 and then the device 10. The sample 51 blocks transmission of the light at the dark portions. Note that each simulation is normalised to its brightest pixel. Each of the simulations (a)-(i) show clear edge enhancement (bright points at the edge, or transition, between blocked and non-blocked portions of the incident light field) — edge enhancement corresponding to the effect of a high-pass filter.

Figure 9 shows an experimentally generated image due to a high-pass filter element 10 as described herein. The figure shows an on-resonance image 60 and two off-resonance images 61. The on-resonance image 60 in shows significant edge enhancement, whereas the off- resonance images (548 nm and 597 nm incident light) show little or no edge enhancement.

The parameter space regarding HRI layer thickness (for a particular HRI material), duty cycle C, period P, and patch element thickness H is complex and not necessarily orthogonal, and therefore it cannot be guaranteed that an approach of maximising performance of parameters individually will produce a maximally optimal device 10 for a particular wavelength. However, the approach may provide a sufficiently useful result, and at least, guidance for further simulation or experimental testing.

Described above is a device 10 used for edge detection. However, other uses are envisaged, which will depend on, for example, the selection of the various parameters described herein. Figure 10 shows an example of a device 10 configured to convert phase gradients into an intensity modulation (such a configured device 10 may be referred to as a phase visualisation device 10). Specifically, a Gaussian shaped phase gradient is converted into an intensity modulation through a device 10 having a period P of 400 nm and illuminated with electromagnetic radiation at a wavelength of 647 nm as a function of maximum phase excursion (cpmax) for unpolarised light (first row), p-polarised light (second row), and s-polarised light (third row). The device 10 of Figure 10 may be suitable, for example, for visualising biological cells which may affect phase substantially more than amplitude (i.e. there is little intensity contrast due to the sample 51).

Figure 20 shows an experimental result for several devices 10 having a grating periods P = 350 nm, P = 400 nm, and P = 450 nm illuminated with plane-wave light sources 50 having different central wavelengths. The column labelled “operational wavelength” shows transmission of the devices 10 for the wavelengths for which the different devices 10 were designed (“on-resonance”). The remaining columns shows the effect of “off-resonance light” sources 50 — those with wavelengths different to that for which each device 10 was designed. As can be seen, the strength of the filter effect decreases away from on-resonance.

Figure 11 shows representative SEM images of devices 10 configured as described herein. Each image is of a device 10 having silver patch elements 13 on a 100 nm TiCh HRI layer 12. The nominal periods P are (a) 350 nm, (b) 400 nm, and (c) 450 nm. Image (d) is of the same device 10 as image (b), tilted by 45 degrees. The duty cycle C is 60%.

Figure 12 shows a plot of reflectance of a device 10 configured in reflection mode, with the following parameters: period P = 400nm; duty cycle C = 60%; patch element height 40 nm, with an incident wavelength of 671 nm.

In an embodiment, a microscope glass slide can be modified to include an electromagnetic filter device 10 according to the embodiments described herein. For example, the device 10 can be affixed to the glass slide. Alternatively, for example, the device 10 can be formed onto the glass slide — for example, the glass slide may correspond to the substrate 11. The microscope glass slide may allow for improved visualisation of live biological cells in real time — that is, the device 40 can provide edge contrast improvements.

In an embodiment, a camera, microscope, or other optical system comprises a filter attachment including a device 10, wherein the filter attachment is attachable to the camera, microscope, or other optical system. For example, the filter attachment can comprise a substrate (such as glass or another material) onto which the device 10 is formed or affixed. In another example, a sticker comprising a device 10 is provided, which for example, can be applied in proximity to an imaging sensor or lens.

Other uses of the devices 10 herein described are envisaged. A particular device 10 can be configured to provide a desired spatial frequency response. For example, a bandpass spatial frequency filter or a low-pass spatial frequency filter.

The embodiments shown herein comprise square or “patch” shaped patch elements 13 (with respect to a plan view). However, other geometries of patch element 13 can be utilised, including mixtures of two or more patch elements 13. For example, Figure 13 shows patch elements 13 having a square-ring geometry and a corresponding plot of transmission for different angles of incident. Also shown is a calculated plot of transmission vs angle of incidence for a device 10 having square-ring patch elements 13. Other possible geometries include rods, circles, and stars.

Figure 15 shows an embodiment of the electromagnetic filter device 10 comprising an array of elongate strip patch elements 43, where the arrangement effectively defines a one dimensional array. As shown, the device 10 also comprises the optional cover layer 13, as well as HRI layer 12 and substrate 11. The operation of the device 10 is similar to the other embodiments described herein, however, the optical transfer function operates predominantly in a direction associated with the one-dimensional arrangement. In Figure 16, an experimental result is shown for such an array having a grating period P = 350 nm illuminated with a light source 50 having a central wavelength of 579 nm. The top illustration shows the Fourier plane images under unpolarised, x-polarised, and y-polarised light — as can be seen, due to the 1- dimensional nature of the array, the x-polarised and y-polarised images are different. Figure 17 shows experimental results of an edge-detection arrangement having a one dimensional array. The results include off-resonance images 71 and on-resonance images 70a- 70c. Of the on-resonance images 70a-70c, included is an image as a result of unpolarised light (image 70a), y-polarised light (image 70b), and x-polarised light (image 70c). The polarisation is defined by reference to the one-dimensional array, such that the y-polarised light interacts with the one-dimensional array and the x-polarised light substantially does not. Therefore, image 70b shows an edge enhancement effect but only with respect to edges parallel the longitudinal axis of the patch elements 13, whereas image 70c shows relatively no edge enhancement. Such an arrangement of patch elements 13 and HRI layer 13 are therefore suitable for enabling increased transmission with respect to one state of polarisation when compared to another state of polarisation of the incident electromagnetic radiation.

In use, an asymmetric optical transfer function may be effectively imposed by tilting the device 10 or sample relative to the incident light (i.e. a tilt with respect to a normal incident of the illuminating light source 50). With reference to Figure 18, for example, a tilt of the device 10 or sample by approximately Q = 2° - 4° (as labelled) relative to the incident beam may effectively modify the optical transfer function such that is shows a near-linear dependence on positive spatial frequencies but little response for small negative frequencies. As shown in the graphs A-C, the intensity of transmitted light becomes asymmetric with respect to positive and negative values of spatial frequency as the device is tilted from 0 = 0 degrees to 3 degrees.

Figure 19 shows experimental results of imaging a sample comprising HeLa cells. The label “NEC” refers to a device 10 comprising a two-dimensional array of patch elements 13 (with reference to Figures 1 and 2). (b) in Figure 16 shows an image of the HeLa cells when the device 10 is utilised without tilt, and (c) shows an image of the HeLa cells when the device 10 is tilted by 3 degrees. As can be seen, significantly improved edge-enhancement is found when the device 10 is tilted, although cell features are identifiable when the device 10 is not tilted. Generally, the usefulness of tilting the device 10 is dependent, at least in part, on the sample being imaged.

Figure 19 also shows the effect of no device 10 present — as can be seen, the cells of the sample cannot be identified (d) and (e) correspond to conventional imaging using (d) Differential Interference Contrast (DIC) and (e) Fluorescent techniques (d) and (e) show that the device 10 provides accurate edge enhancement when viewing the cell samples. The inserts at the bottom left show enlarged detail of features 1 and 2 of the corresponding images (c), (d), and (e) (when read left to right). The top-left insert illustrates the experimental setup.

Figure 14 shows an example method for manufacturing a device 10. At step 100, desired device 10 parameters are determined, for example according to a simulation as described herein. Then, at step 101, a selected HRI layer 12 is formed onto a substrate 11 (such as a glass substrate) to a desired thickness. For example, a TiCh layer can be evaporated on top of the substrate 11 via e-beam evaporation to a desired thickness (e.g. 100 nm).

At step 102, a resist (e.g. a PMMA EBL resist) is spun onto the top of the HRI layer 12 and at step 103 the resist is developed (e.g. using EBL) to inscribe the arrangement of patch elements 13 onto the surface of the HRI layer 13. At step 104, a desired thickness (e.g. 40 nm) of silver is evaporated onto the top surface, for example, using e-beam evaporation. Step 105 is a lift-off step of the silver layer. Finally, the device is covered with a cover layer 14, such as an approximately 0.75 - 1 pm PMMA layer, at step 106. The PMMA cover layer 14 may be formed by being spun and baked at 180 °C for 3 minutes. In a variation, a glass, SiCh, or other layer is evaporated or otherwise formed onto the HRI layer 13, thereby producing a cover layer 14.

The device 10 may be configured for interaction with one or more different polarisations of light. The device 10 may be configured to show different interactions with differently polarised light. For example, the one-dimensional array of patch elements 13 interacts differently with x- and y- polarised light. More generally, the device 10 may interact with orthogonally polarised light different — for example, oppositely circularly polarised light.

According to an embodiment, the arrangement of patch elements 13 comprises a varying spacing between the patch elements 13. For example, a sinusoidal-like variation in spacing may broaden the range of incident wavelengths of light for which the device 10 may be suitable.

Further modifications can be made without departing from the spirit and scope of the specification. For example, additional control of the phase and resonances within the device can be achieved by utilising, as patch elements 13, with plasmonically-active nanostructures. In another example, the patch elements 13 may comprise a blaze profile.

Claims

CLAIMS:

1. An electromagnetic filter device comprising a substrate and a high refractive index dielectric layer (HRI layer) located on a top surface of the substrate, and an arrangement of patch elements extending past a top surface of the HRI layer, wherein the patch elements are arranged to cause a coupling of incident electromagnetic radiation to the HRI layer, and wherein the HRI layer is configured to directly couple electromagnetic radiation at each patch element to one or more other patch elements.

2. An electromagnetic filter device as claimed in claim 1, wherein the arrangement of patch elements is configured to implement a corresponding optical transfer function with respect to light incident at a preselected angle.

3. An electromagnetic filter device as claimed in claim 2, wherein the arrangement of patch elements is configured to implement a high-pass spatial frequency filter.

4. An electromagnetic filter device as claimed in any one of claims 1 to 3, wherein the patch elements are arranged with a constant period in at least one direction, wherein the period is selected based on at least an intended wavelength of the electromagnetic radiation.

5. An electromagnetic filter device as claimed in claim 4, wherein the patch period is smaller than the wavelength of the intended electromagnetic radiation.

6. An electromagnetic filter device as claimed in any one of claims 1 to 5, wherein the HRI layer comprises TiCh.

7. An electromagnetic filter device as claimed in any one of claims 1 to 6, wherein the patch elements comprise a metal or a dielectric.

8. An electromagnetic filter device as claimed in any one of claims 1 to 7, wherein the patch elements are formed on top of the HRI layer.

9. An electromagnetic filter device as claimed in any one of claims 1 to 8, further comprising a cover layer applied to the top layer of the HRI layer such as to cover the patch elements, wherein the refractive index of the cover layer is lower than the refractive index of the HRI layer.

10. An electromagnetic filter device as claimed in claim 9, wherein the cover layer comprises polymethyl methacrylate (PMMA), a glass, or SiC .

11. An electromagnetic filter device as claimed in any one of claims 1 to 10, wherein the patch elements are arranged in a regular array.

12. An electromagnetic filter device as claimed in claim 11, wherein the regular array is a two-dimensional rectangular array.

13. An electromagnetic filter device as claimed in claim 11, wherein the regular array is a one-dimensional array comprising elongate path elements.

14. An electromagnetic filter device as claimed in any one of claims 1 to 13, wherein the arrangement of patch elements and HRI layer are configured such that the electromagnetic filter device shows increased transmission with respect to one angle of incidence when compared to another angle of incidence of the incident electromagnetic radiation.

15. An electromagnetic filter device as claimed in any one of claims 1 to 14, wherein the arrangement of patch elements is configured to interact with a first polarisation of incident light differently to a second polarisation of light.

16. An electromagnetic filter device as claimed in claim 15, configured to show a higher transmission of the first polarisation when compared to the second polarisation, and wherein the first polarisation is orthogonal to the second polarisation.

17. An electromagnetic filter device as claimed in any one of claims 1 to 16, wherein the substrate comprises a glass, SiCh, or polymethyl methacrylate (PMMA).

18. An electromagnetic filter device as claimed in any one of clams 1 to 17, wherein the device is configured for use with a biological sample.

19. An electromagnetic filter device as claimed any one of claims 1 to 18, wherein the device is integrated with a glass microscope slide.

20. An electromagnetic filter device as claimed in any one of claims 1 to 18, adapted for use as an optical component of a camera.

21. A method of imaging a sample using an electromagnetic filter device according to any one of claims 1 to 20, comprising positioning the sample between an electromagnetic radiation source and the electromagnetic filter device, illuminating the sample with electromagnetic radiation from the electromagnetic radiation source, and capturing an image of the electromagnetic radiation transmitted through the electromagnetic filter device.

22. A method as claimed in claim 21, wherein the sample is a biological sample.

23. A filter attachment for a camera, microscope, or optical system, wherein the filter attachment comprises the electromagnetic filter device of any one of claims 1 to 18.

24. A camera, microscope, or optical system comprising the electromagnetic filter device of any one of claims 1 to 18.