GB2587024A - A sensor - Google Patents

Abstract

Optical waveguide 2 is formed on the surface of transparent substrate 3. Light source 6-10 injects convergent beam 100 having curved wave-fronts into the waveguide through its interface with the substrate. Detector 5 receives light in a range of angles returned from the waveguide via the substrate, and detects a change in angle of a peak or trough 11 of intensity (maybe a dip-peak pair). It thereby determines the presence of a substance in contact with the waveguide, by determining a change in refractive index at the substance/waveguide interface. The waveguide may support leaky modes, and may be porous to admit diffusion of the substance into the waveguide. The sensor may be a pH sensor, wherein pH changes cause a hydrogel to swell or shrink; it may be a chlorine sensor, wherein a hydrogel matrix contains chlorine-reactive groups; it may be a biosensor, wherein biological analyte binds to a hydrogel matrix.

Description

A SENSOR

FIELD

[1] The invention relates to sensors and methods of sensing, such as optical sensors and optical sensing methods. In particular, though not exclusively, the invention relates to leaky waveguide optical sensors for detecting a substance placed in contact with the sensor, according to changes in the optical properties of the sensor induced by the substance.

BACKGROUND

[2] Optical waveguide sensors provide a mechanism by which light guided within a waveguide, between the internal surface of the waveguide, is able to interact with a substance placed in contact with an external surface of the waveguide. The waveguide is structured in such a way that the evanescent electrical field of the guided light wave is able to penetrate into the external substance through the interface formed between the waveguide and the external substance. The interaction of the penetrating evanescent field with the external substance influences the optical properties (e.g. intensity) of the light guided within the waveguide. Upon subsequently extracting the guided light, measured changes in these optical properties may be found to correlate with changes in properties of the substance.

[3] In order to maximise the extent and penetrative power of the evanescent field of the guided light, existing methods apply a coating of a metal such as gold, deposited on a surface of the waveguide opposite to the one with which the substance makes contact. The observed intensity changes in the guided light (e.g. dips or peaks in reflectivity) then depend on the complex refractive index of the metal. However, vacuum deposition is required for the fabrication of a sufficiently high quality metal layer when manufacturing such optical waveguide sensors, and this adds significant complexity and cost to the manufacturing process.

[4] The invention aims to provide an alternative. SUMMARY [5] At its most general, the invention is an optical sensor based on the direct observation of dip-peak pairs in the light reflected from an optical waveguide, such as a so-called leaky waveguide (LVV). This phenomenon is unexpected and is found to be increasingly apparent when the refractive index contrast between the waveguide material and a sample material becomes lower. For example, a refractive index contrast equal to or less than about 0.00500 is preferable, such as a contrast in the range of about 0.00200-0.00500 for example. The waveguide may be illuminated obliquely through a side of the waveguide, with a beam of light configured to inject light into the waveguide covering a range of angles of incidence simultaneously. A convergent beam of light such as a wedge-shaped beam may be used for this purpose. The waveguide is believed to act as a Gires-Tournois optical resonator at particular angles of incidence of the beam of light at which the waveguide resonates to support resonator modes in the waveguide. This is believed to result in a coupling between transverse modes and longitudinal modes of the light beam and the resulting formation of dip-peak pairs for each waveguide mode.

[6] The waveguide may provide sensitivity comparable to that of surface plasmon resonance (SPR) optical sensors, but without the requirement for a metallic coating essential for SPR sensors, as described above. The waveguide may comprise a Hydrogel Optical Waveguide (HOVV). This may comprise a slab waveguide. The waveguide may be disposed on a solid substrate (e.g. glass), for support. The refractive index of the material of the LW may be intermediate between that of the substrate and the sample. A process other than total internal reflection (TIR) may occur within the waveguide to induce limited confinement of light in at the waveguide making the waveguide leaky, leading to a coupling of light into and out of the waveguide at specific resonance angles of oblique incidence.

[7] The observation of dip-peak pairs in the waveguide reflectivity curve occurs strongly for very low refractive index waveguides (or low refractive index contrast with a sample material in contact with it), allowing the visualisation of the appearance of waveguide modes at appropriate resonance angles. Illuminating the waveguide with light at resonance angles of incidence has been found to cause the reflected wavefront to contain a sharp 2Tr phase change at the resonance angle. This sharp phase change is believed to signal the occurrence of an optical resonator mode analogous to that in a Gires-Tournois optical resonator. The use of a light beam with a curved optical wavefront is believed to promote an optical coupling between transverse modes and longitudinal modes of the light beam to create resonance modes in the resonator, resulting in a dip followed by a peak in reflectivity. This phenomenon has is surprising. The dip-peak pairs were also not seen using transfer matrix mathematical modelling using models that assumes a plane input wavefront. Conversely, models assuming a curved convergent wavefront did show the dip-peak pairs. This is confirmed experimentally. A low refractive index contrast between the waveguide and the sensed sample in contact with it is found to significantly support the strong dip-peak features in the reflectivity curve at resonance angles.

[8] In a first aspect, the invention may provide a sensor for sensing a change in a refractive index caused by the presence of a substance, comprising: an optically transparent substrate (e.g. a surface of a prism); an optical waveguide formed upon a surface of the substrate thereat to form a first optical interface with the surface of the substrate for receiving light therethrough, wherein the optical waveguide comprises an external surface opposing the first optical interface for forming a second optical interface by contact with said substance; a light source configured to inject into the optical waveguide a convergent beam of probe light having a curved optical wavefront such that the probe light is guided along the optical waveguide between the first and second optical interfaces; a detector configured to receive probe light returned from the waveguide via the first optical interface across an interval of oblique angles (concurrently or successively) relative to the first optical interface; wherein the detector is arranged to detect a change in an angle within said interval at which the received probe light achieves a peak or a trough in light intensity, and to determine a change in the refractive index at the second optical interface caused by the presence of the substance according to the change in said angle.

[9] In this way, the optical modes and the resonances set up in the waveguide by the injection of the probe light into it, are found to be sensitive to the change in refractive index at the second optical interface. This change in refractive index may result from a change in the refractive index of the sample material in contact with the second optical interface externally, and/or may also result from an ingress of sample material into the body of the optical waveguide (which may be porous) so as to change the effective refractive index of the optical waveguide itself at the second optical interface or/and at depths below that interface and within the body of the optical waveguide.

[10] The optical waveguide may be configured for use in detecting changes of refractive index caused by the presence of a substance within a fluid, such as a gas or a liquid (e.g. aqueous) solution. This is advantageous also if the waveguide is porous and intended to absorb the fluid into its porous structure as a part of the detection process. It is found that the use of an optical waveguide having a refractive index close to (e.g. greater than) that of the fluid is beneficial in supporting strong dip-peak structures in the angular spectrum of the reflection spectrum of the waveguide as detected by the detector.

[11] The optical waveguide may be formed from a material having a refractive index not exceeding 1.38000 in respect of light having a wavelength of 650nm. More preferably, the optical waveguide may be formed from a material having a refractive index not exceeding 1.36800 in respect of light having a wavelength of 650nm. For example, the optical waveguide may be formed from a material having a refractive index on the range from 1.30000 to 1.36800 in respect of light having a wavelength of 650nm.

[12] The optical waveguide is most preferably a leaky waveguide (LVV) configured to support leaky optical modes in respect of the injected probe light. Accordingly, by using a LW the out-coupling of light from within the waveguide, at the resonance angles of the waveguide, is made very efficient and improves sensitivity.

[13] The optical waveguide is preferably porous. The optical waveguide preferably comprises a porous material adapted to admit diffusion of an aforesaid substance into the optical waveguide via the second optical interface. As discussed above, this is desirable to permit the effective refractive index of the body of the optical waveguide to change in response to the presence of a substance being (to be) detected. The porosity may be configured to admit diffusion of liquids, or gasses, for the purposes of inducing/causing a change in the effective refractive index of the optical waveguide itself.

[14] The optical waveguide may comprise a material forming a hydrogel matrix. The optical waveguide may be formed from a hydrogel or a dehydrated hydrogel matrix which is rehydratable to form a hydrogel.

[15] The light source may be arranged to form a convergent beam of probe light which is convergent in all dimensions (e.g. x-direction; y-direction) transverse to the longitudinal/propagation axis (e.g. z-direction) of the beam. For example, the beam may be conically convergent, with a circular or elliptical cross-sectional shape.

[16] The light source is preferably arranged to form said convergent beam of probe light as a wedge-shaped beam which is convergent in a first direction (e.g. x-direction) orthogonal to the axis of the beam of probe light (e.g. z-direction of propagation), but which is not convergent (or is less convergent) in a second direction (e.g. y-direction)which is orthogonal to both the first direction and said axis. For example, the beam may be convergent, with a rectangular cross-sectional shape. The direction of non-convergence (of lesser convergence) may preferably be parallel to the surface of the waveguide, and this may promote transverse modes of the beam parallel to the direction of beam convergence and transverse to the surface of the waveguide, for coupling to longitudinal beam modes reflecting back and forth between opposing waveguide surfaces within the waveguide. This may assist in generating Gires-Tournois resonator-like resonances within the waveguide.

[17] The optical waveguide may be a planar waveguide in which the first optical interface and the second optical interface are disposed in substantially plane-parallel relative opposition and separated by a substantially uniform waveguide thickness of at least 1pm. The waveguide thickness preferably does not exceeding lOpm.

[18] The detector may be arranged to determine: a first aforesaid angle at which the received probe light achieves a dip or trough in light intensity; a second aforesaid angle at which the received probe light achieves a peak in light intensity; a mean as an average of the first aforesaid angle and the second aforesaid angle; a change in the refractive index at the second optical interface (e.g. of the substance and/or of the waveguide) according to a change in the mean angle.

[19] The second angle is preferably greater than the first aforesaid angle.

[20] The peak is preferably a maximum (e.g. the greatest value achieved) in the light intensity within the interval of angles. The trough is preferably a minimum (e.g. the lowest value achieved) in the light intensity within the interval of angles.

[21] The peak is preferably the first peak that follows an aforesaid trough in angular distance along the interval of angles. For example, the peak may be paired with the dip/trough.

[22] In a second aspect, the invention may provide a pH sensor comprising a sensor as described above in which the optical waveguide comprises a material forming a hydrogel matrix adapted to swell or shrink thereby to change the refractive index of the optical waveguide in reaction to changes in the pH of said substance at said second optical interface.

[23] In a third aspect, the invention may provide a chlorine sensor comprising a sensor as described above in which the optical waveguide comprises a material forming a hydrogel matrix containing chlorine-reactive groups adapted to change the refractive index of the optical waveguide in reaction to chlorine at said second optical interface.

[24] In a fourth aspect, the invention may provide a biosensor comprising a sensor as described above in which the optical waveguide comprises a material forming a hydrogel matrix adapted to permit the binding thereto of biological analyte materials via said second optical interface. The binding of an analyte to bio-recognition species immobilised in the waveguide changes its refractive index and the angular position of the dip-peak pair, which provides the bio-sensing mechanism.

[25] It is to be understood that the sensor described above implements a corresponding method for sensing a change in a refractive index. The method is not intended to be limited to implementation by a particular apparatus and, for completeness, the corresponding method is detailed below.

[26] In a fifth aspect, the invention may provide a method for sensing a change in a refractive index cause by the presence of a substance, comprising: providing an optically transparent substrate bearing an optical waveguide upon a surface of the substrate defining a first optical interface with the surface of the substrate for receiving light therethrough; by contact with said substance, forming a second optical interface at an external surface of the optical waveguide opposing the first optical interface; injecting into the optical waveguide a convergent beam of probe light having a curved optical wavefront such that the probe light is guided along the optical waveguide between the first and second optical interfaces; receiving probe light returned from the waveguide via the first optical interface across an interval of oblique angles (e.g. concurrently or successively) relative to the first optical interface, detecting a change in an angle within said interval at which the received probe light achieves a peak or a trough in light intensity, and determining a change in the refractive index at the second optical interface caused by the presence of the substance according to the change in the angle.

[27] The method may include forming the convergent beam of probe light as a conical beam of light or as a wedge-shaped beam of light which is convergent in a first direction orthogonal to the axis of the beam of probe light, but which is not convergent in a second direction which is orthogonal to both the first direction and that axis.

[28] The method may include the steps, performed by the detector, of determining: a first said angle at which the received probe light achieves a trough in light intensity; a second said angle at which the received probe light achieves a peak in light intensity; a mean angle as an average of the first angle and the second angle; and a change in the refractive index at the second optical interface (e.g. of the substance and/or the optical waveguide) according to a change in said mean angle.

[29] The optical waveguide may be formed from a material having a refractive index not exceeding 1.38000 in respect of light having a wavelength of 650nm. More preferably, the optical waveguide may be formed from a material having a refractive index not exceeding 1.36800 in respect of light having a wavelength of 650nm. For example, the optical waveguide may be formed from a material having a refractive index on the range from 1.30000 to 1.36800 in respect of light having a wavelength of 650nm.

[30] The optical waveguide is most preferably a leaky waveguide configured to support leaky optical modes in respect of the injected probe light.

[31] The optical waveguide is preferably porous. The optical waveguide preferably comprises a porous material adapted to admit diffusion of an aforesaid substance into the optical waveguide via the second optical interface.

[32] The optical waveguide may comprise a material forming a hydrogel matrix. The optical waveguide may be formed from a hydrogel or a dehydrated hydrogel matrix which is rehydratable to form a hydrogel.

[33] The optical waveguide may be a planar waveguide in which the first optical interface and the second optical interface are disposed in substantially plane-parallel relative opposition and separated by a substantially uniform waveguide thickness of at least 1 pm. The waveguide thickness preferably does not exceeding lOpm.

[34] The second angle is preferably greater than the first aforesaid angle.

[35] The peak is preferably a maximum in the light intensity within the interval of angles. The trough is preferably a minimum in the light intensity within the interval of angles.

[36] The peak is preferably the first peak that follows an aforesaid trough in angular distance along the interval of angles.

[37] The method may be a method for pH sensing in which the optical waveguide comprises a material forming a hydrogel matrix adapted to swell or shrink thereby to change the refractive index of the optical waveguide in reaction to changes in the pH of said substance at said second optical interface.

[38] The method may be a method sensing chorine in which the optical waveguide comprises a material forming a hydrogel matrix containing chlorine-reactive groups adapted change the refractive index of the optical waveguide in reaction to chlorine at said second optical interface.

[39] The method may be a method for sensing a biological analyte material in which the optical waveguide comprises a material forming a hydrogel matrix adapted to permit the binding thereto of biological analyte materials via said second optical interface.

BRIEF DESCRIPTION OF DRAWINGS

[40] There now follows a description of illustrative examples of the invention with reference to the accompanying drawings, of which: [41] Figure 1 shows a schematic diagram of a sensor apparatus; [42] Figure 2 shows a diagram of a sensor apparatus; [43] Figure 3A shows a schematic diagram of a Gires-Tournois (GT) optical resonator; [44] Figure 3B shows a schematic diagram of a reflectivity and an optical phase shift of reflected light from the GT resonator of Figure 3A, as a function of light incidence angle; [45] Figure 3C schematically shows a wedge-shaped optical beam convergent upon a leaky waveguide obliquely; [46] Figure 4 shows an optical reflectivity (and phase shift: see inset) angular resonance spectrum of light out-coupled from a leaky waveguide (acting as a GT resonator) according to an example of the invention, together with a schematic angular spectrum indicating the angular distribution of GT resonator modes caused by an optical coupling between longitudinal modes and transverse modes of the probe beam having a curved optical wavefront; [47] Figure 5 shows the calculated optical diffraction spectrum of light out-coupled from a leaky waveguide according to a mathematical model consistent with an example of the invention; [48] Figures 6(a) and 6(b) show the calculated optical diffraction spectra of light out-coupled from a leaky waveguide according to a mathematical model consistent with an example of the invention; [49] Figure 7 shows a diagram illustrating changes in the detected angular position of an angle at which probe light out-coupled from a leaky waveguide achieves a peak or a trough in light intensity, according to an example of the invention; [50] Figure 8 shows a further diagram illustrating changes in the detected angular position of an angle at which probe light out-coupled from a leaky waveguide achieves a peak or a trough in light intensity, according to an example of the invention.

DESCRIPTION OF EMBODIMENTS

[51] In the drawings, like items are assigned like reference symbols.

[52] Figure 1 shows a schematic diagram of a sensor apparatus 1 according to an embodiment of the invention. Figure 2 shows an apparatus consistent with the schematic diagram of Figure 1. The sensor apparatus comprises a planar leaky waveguide (LVV) 2 formed from a hydrogel strip or slab of uniform thickness supported upon the plane surface of a glass prism 3. The glass prism, supported within a prism holder 3A, has a refractive index (at optical wavelengths of light) exceeding that of the LW and provides a substrate forming a first optical interface through which probe light 100 from a light source (items 6 -10) is coupled into the LW.

[53] A flow cell 4 is disposed at the upper, exposed surface of the LW opposite to the first optical interface and, in conjunction with the exposed surface of the LW, it defines a fluid conduit for directing a flow of a fluid substance from a flow cell fluid inlet 13, across the exposed surface of the LW and to a flow cell outlet 14. Whilst flowing across the exposed surface of the LW in contact with it, the substance forms a second optical interface with the LW.

[54] The light source comprises a light emitter 10, such as an LED, which may be monochromatic but need not be, and an optical train (items 6-9) for collecting light emitted by the light emitter, collimating it into a collimated light beam, linearly polarising the light of the collimated light beam, and subsequently shaping the polarised and collimated light beam into a beam shape which converges in only one dimension (orthogonal to the longitudinal beam axis) to define a wedge shaped beam. The optical train comprises an achromatic lens doublet 9 for collecting and collimating the light emitted by the light emitter, followed by a linear polarising filter 8 disposed to receive the collimated light beam and to transmit the portion thereof which is TE linearly polarised, and a subsequent cylindrical lens 6 and subsequent aperture stop 7, configured to receive the TE polarised beam of collimated light and re-shape it converge into a wedge shape 100 (see Figure 3C) the converged wedge tip of which resides within the LW and extends in a direction coplanar with the plane of the LW. The wedge angle is -14° in air or -9° in the prism.

[55] The light source is disposed to direct the wedge-shaped light bean in to the LW via an input face of the coupling prism 3 in a direction generally or approximately perpendicular to the input face, and at an input angle (0) coinciding with an angle at which the LW is expected to support leaky waveguide modes for the injected light of the beam. As shown in Figure 2, an achromatic doublet lens 12 is optionally mounted on a rotating stage together with a camera 5 and configured to focus the reflected probe light beam 101 (containing dip-peak structures 11) upon the sensor chip of the camera 5. Rotation of the rotation stage permits the doublet lens and camera to be rotated, in unison, to an appropriate angular position to receive the reflected probe light. In addition, the camera 5 is mounted upon a linear translation stage permitting translation of the photo-detector of the camera in scanning direction which is transverse to the longitudinal axis of the reflected wedge beam 101 and transverse to the length of the converged wedge tip (see 102: Fig.3C) of which resides within the LW. This permits the dip-peak structures 11 to be positioned upon the camera sensor chip as appropriate. The rotational and linear translation stages, and the achromatic doublet lens 12, may be omitted if desired, as shown in Figure 1.

[56] The wedge-shaped nature of the wedge beam ensures that LW is concurrently illuminated by light possessing a curved and convergent optical wavefront spanning an interval of incidence angles. Reflection and diffraction of light of the wedge-shaped beam occurs and returns a reflected wedge of light 101 (see Figure 3C) through the first optical interface in a direction back into the coupling prism 3 towards the camera 5 disposed and at an output angle corresponding with the angle (e) of incidence of the beam of light, in the plane containing the input angle and the normal to the first optical interface. The area of the photo-sensor chip of the camera is sufficient to capture a cross-section of the output wedge beam containing not only the reflected probe light 100 and the central un-diffracted light, but also concurrently containing higher orders of diffracted light defining peak-dip fringes 11.

[57] In alternative arrangements, a scanning photodetector may be used which is configured to detect the reflected/out-coupled probe light over only a narrow angular range not encompassing the full angular range of a group of resonances, but which is configured to scan across successive angles to cover the entire angular range of the resonances and to detect/record reflectivities at all angles within the range scanned, successively rather than concurrently.

[58] Direct visualisation of LW modes: Resonance and diffraction [59] Gires-Toumois Optical Resonators [60] It is useful for an understanding of the invention to consider a Gires-Tournois optical resonator. As shown schematically in Figure 3A, these comprise two typically parallel reflecting surfaces: an upper reflector which is partially reflecting (e.g. with reflectivity, r1<1) and a lower reflector which is highly reflecting (e.g. with reflectivity r2r-1) and is separated by a uniform distance (d) from the upper reflector. Light is injected, as a beam, into the resonator through the upper reflector at an input angle (6), and is reflected repeatedly between the lower reflector and the upper reflector, internally, many times passing through the material within the resonator cavity of refractive index n (e.g. glass, with r1.5). Upon each reflection from the upper reflector, internally, some of the light is reflected back in to the resonator again, while some is transmitted through the reflector as an output.

[61] Interference between multiply-reflected beams within the resonator results in destructive optical interference at certain optical path geometries within the resonator (between the upper and lower reflecting surfaces). Variation in the optical path geometry may be achieved by varying the wavelengths, A, of light input into the resonator, at a fixed input angle, 0, or by varying the input angle, 0, for a fixed wavelength of light, A. The result is an amplitude reflectivity of the resonator given by: ri-e ta = 1-r, eth Here, 6 is a phase given by the geometry of the resonator: 4Thrld = -cos(0) 2 [62] Multiple reflections of the light within the resonator cavity causes a non-linear phase in the output light transmitted from the cavity through its upper surface. A schematic example of the non-linear output phase-shift of the output light is shown in Figure 3B, and the reflectivity of the resonator is shown in Figure 3B.

[63] This non-linear output phase shift is given by: = 2 tan-1 f 1 + a tan (5)} 1. 1-VT W) [64] Here, R = , which is the intensity reflectivity of the upper reflector. This non-linear phase shows a step-like shape, in steps of size 2-rr, as a changes (i.e. as U changes). If r2becomes smaller than one (e.g. 0.95) when niS low (e.g. 0.3) then the reflected intensity output of the resonator develops a series of intensity dips located at specific resonance angles, 0, of the injected beam of light at which optical resonator modes are supported by the resonator cavity. This signifies retention of some of the optical energy of the beam within the resonator cavity at such resonance angles.

[65] The reflected intensity spectrum of the CT etalon, as well as the phase spectrum, is qualitatively very similar to the reflectivity spectrum observed in the present invention when illuminated by a collimated mechanical scan. Both show intensity 'dips' at certain resonance angles of incidence. However, it is believed that the reflectivity dips observed when mechanically scanning, in angle of incidence, a collimated (i.e. non-converging) probe light beam across the waveguide of the invention, is dominated by scattering losses of light within the waveguide (e.g. porous in structure and prone to some internal light scattering) due to a greater proportion of the light being partially confined within the waveguide for longer (i.e. longer optical path length within the waveguide) when CT resonances occur at CT resonance angles. This is analogous to the losses of light from the schematic CT resonator of Figure 3A, via its rear reflecting surface. Both scenarios result in a loss of light from the CT resonator and a consequent dip in the reflectivity at the CT resonance angle in question.

[66] However, when the incident beam is a wedge beam, the reflectivity spectrum develops much more complex resonance structures comprising series of resonance dip-peak structures. The initial resonance dip of a dip-peak structure coincides with the 'normal' GT resonance dip position, but is joined by a series of closely-spaced smaller resonance dips, and with resonance peaks in between them.

[67] The following discussion aims to provide a qualitative, and purely heuristic understanding of why it is that a wedge beam can induce these new structures, and aims to provide a conceptual basis for concluding, in a much more general way, that the important feature of the wedge beam is that it contains 'transverse modes' within the optical beam. This means that, in principle, any beam shape with sufficient 'transverse modes' could be used in the invention to induce the required dip-peak resonance structures.

[68] Transverse Modes [69] In geometrical optics, a bundle of parallel rays can be focussed to a point by using a thin lens. In reality, a focal spot has a finite size and the effects of diffraction, which are ignored in the geometrical optics regime, are important. By taking account of the wave nature of the light converging to a focus, we illustrate the effect that this may have on the dynamics of the light converging to at a focus in the context of the present invention. To achieve a better understanding of the invention, consider a light beam in which the distribution of light intensity at planes normal to the beam propagation direction is Gaussian in form, and the wavefronts are spherical in curvature. This form is one of those most widely encountered in optical electronics.

[70] Importantly, the Gaussian representation of a light beam forms a complete basis set which can be used to accurately represent many other beam wave functions: i.e. by representing the beam wave function as a weighted sum of Gaussian beams. Thus, gaining an understanding of the properties of a Gaussian beam allows a heuristic understanding of light beams more generally.

[71] When propagating in a homogeneous medium, it can be shown that the electric field of a Gaussian beam propagating along the 'z' axis, is described by: ( El = E0 (c) Hi '7c-/7 H", j'' x exp x2 ± y2 ik x2 + y2. . ikZ ± la ± M ± 1))/) 0)(Z) \w (z)) (0(Z) (02 (z) 2R(z) [72] Here, Fix is a Hermite polynomial of order x, and: w2(z) = 4(1+ (z/z0)2) R(z) = z(1 + (z0/z12) [73] The term m(z) represents the radius of the waist of the light beam at a given point, z, along the axis of propagation. The quantity w, is the radius of the focal spot of the beam, and R(z) is the curvature of the wavefront of light at an axial position z along the z-axis of propagation of the beam. It is noted that the curvature is negative when z<0 (i.e. to the left' of the focus), is infinite at the focus (i.e. becomes planar at that point) and is positive when z>0 (i.e. to the 'right' of the focus). The term q is interpreted as a phase difference between the Gaussian beam and a plane wave of the same wavelength also travelling in the +z direction. The transverse variation, in the x-dimension and the y-dimension, of the electric field of the light beam is therefore Gaussian in profile and of the form: ( X-V) X2) tow) x exp (02 (z) this being in the x-dimension, and: (w(z)) x exp(wy2 (20) this being in the y-dimension. A phase factor is created by the presence of a curvature, R(z), of the wavefront: (x2+ y2 ik 2R(z) ) [74] Note that the phase shift of this Gaussian wave, along the z-axis of propagation, is given by: (pun = kz -(1+ m + 1) tan -1(1) z" [75] If the wave is propagating within an optical resonator, such as Gires-Tournois resonator, with a first of the two resonator reflecting surfaces located at z = ziand a second of the two resonator reflecting surfaces located at z = z2, such that d = z2 -z1 is the width of the resonator, then the condition required for optical resonance to occur within the resonator, is: Sol,m(z3-Sol,m(z2) = Thr [76] Here N is an integer. This gives, for a given fixed value of N: kd -(1 + in. +1)[tan-1 (-z2 -tan' = N7 z" z" Here, the value of k satisfies this resonance condition, and represents a longitudinal resonance mode. Note that the values of the transverse mode indices, 1 and in, within this equation may individually vary without changing the value of (/ + m). Thus, the resonant values of the wave vector k depend on the sum 1+ m and not on the values of the mode indices, 1 and m, individually, meaning that for a given longitudinal resonator mode (i.e. a given value of N), there exist additional degenerate transverse modes with mode indices, 1 and in, constrained such that the sum (/ + m) is fixed.

[77] In the above analysis, we have implicitly assumed that the Gaussian beam is incident to the resonator reflecting surfaces in a direction perpendicular to them. However, an arrangement in which the light rays forming the Gaussian beam enter into the optical resonator at a finite angle of incidence (0), relative to the internal normal of the resonator reflecting surfaces (as in a Gires-Tournois resonator), the above resonance condition applies to the component, k, = kcos(0), of the wave vector k which is projected on to this normal within the resonator. Consequently, for a given fixed value of N associated with a longitudinal mode with a wave vector k, a resonance condition exists for transverse modes with mode indices, 1 and in, associated with that longitudinal mode. In particular, consider the resonance condition for two different values of the sum (1 + m), and correspondingly different values of k kcos (0 (2))d -(I + in ± 1)(2) [tan' (2 -tan-1)[ = NIT Zo Zo z, kcos (0 (1))d -(1 + in + 1)(1) [tan-1 (N -tan' = NTh Zo ZO [78] One can see that it is possible for both equations to be satisfied simultaneously such that 0(2) # 0 (1)provided that (1 + in + 1)(2) # (1 + m + 1)0) in a compensating way. The effect of the difference in the angles 0, is compensated for by a corresponding difference in the transverse mode indices (I + in). Taking the difference between the above two equations gives: (cos (0(2)) -cos (0(1)))kd = [(I + in + 1)(2)-(i in + 1)1 x [tan' -tan' ()1 zo zo [79] For constant resonator dimensions, beam dimensions and wavelength of light we have: cos(0 (2)) -cos(OW) = [(1 + m + 1)(2)-(7 + m + IP)] x Constant In this sense, the resonance condition for a given value of the wave vector k satisfying the longitudinal resonance mode condition in the resonator, can be simultaneously satisfied by coupling light (i.e. diffracting within the beam) into to multiple corresponding transverse modes with mode indices, 1 and m, at different corresponding angles of incidence of the converging light rays forming the beam.

[80] For two closely-spaced plane reflecting surfaces of a resonator in which Z2 and z, and small compared to z0, we may approximate: aQ co s (0 (2)) -cos (0°)) - + in) kz, If 0(2) 74--; 0(0, then using the 'cosine difference rule' gives: A(/ + in) (0(2) _ (1)) Here, A(1+ m) E [(I + 777 1)(2)-(i in + 1)(1.

kzo sin(OW) [81] Whereas the angular positions of successive longitudinal resonator modes (N, N+1, N+2...) will be more widely-spaced in angular separation, with angle cosines differing by 771 kd, each will have associated with it a set of more closely-spaced transverse mode resonances, A(1+ in) = 1,2,3, etc, given that zo >> d.

[82] If transverse modes (/ + TO are absent from the light beam, such as if there is no curvature in the wavefronts of the beam, e.g. caused by focussing, then these additional resonances will be absent from the resonator resonance pattern: only longitudinal resonance modes will be present. The act of forming a convergent optical beam creates transverse modes within the beam and allows these pre-existing transverse modes to couple to longitudinal modes in the resonator cavity thereby to permit additional resonance modes within the resonator cavity, when the light beam is injected into it. The transverse modes of the beam are associated with variations in the electrical field of the beam along axes (x, y) transverse to the propagation axis (z), and one of these axes may correspond to a direction extending across (e.g. plane parallel) to the plane of the reflecting surfaces of the optical resonator. By forming an extended focal region (e.g. a wedge beam) with a focal line parallel to this transverse axis, one may sympathetically generate transverse beam modes in this dimension.

[83] Qualitatively, this is observed in the reflectivity spectrum of the present invention, as shown in Figure 4, indicating that the LW functions, qualitatively, as a GT optical resonator cavity. Figure 4 shows the output of the LW of the embodiment of the invention when illuminated through a prism with a converging wedge beam (i.e. a range of angles of incidence simultaneously) produced by a cylindrical lens. The dip-peak resonance structure created in the reflected (out-coupled) probe light beam is imaged by a camera, and an image of the result is provided in Figure 4 inset (i) along with the corresponding intensity profile as a function of varying probe beam angle of incidence (and angle of reflection, by symmetry). The intensity profile of the same LW when illuminated with a mechanically scanned collimated beam and recorded by a photodiode is also shown in Figure 4.

[84] The reflectivity curve obtained by the mechanical scan of the collimated beam shows the presence of dips corresponding to the resonance angles. These dips in the reflectivity curve were quite shallow and a result of small scattering losses in the poly(AAm) waveguide. In contrast, the reflectivity curve obtained by illuminating the LW with a wedge-shaped beam resulted in dip-peak pairs at the resonance angles. Figure 4 shows that the angular width of the dip-peak pair corresponding to the zero-order LW mode was narrowest. This observation can be explained by considering a plot of the phase of the reflected wavefront versus the angle of incidence (inset (ii), item 43 in Figure 4), which illustrates that the 2-rr phase change for the zero-order LW mode was steepest. Additionally, the depth/ height of the dip/ peak decreased as the order of the LW mode increased because the zero-order optical mode is better confined in the waveguide than the second-order mode. The resonance angle corresponding to the zero-order mode is greater than the higher-order modes. As discussed previously, the Fresnel reflectivity at the substrate/waveguide interface increases at high angles, thus increasing the Q factor and in turn leading to improved confinement of the zero-order mode in the waveguide.

[85] Figure 4 shows that the reflectivity 40 of the LW across a range of angles of incidence of light from the probe light beam 100, encompassing several of the mode resonances imaged simultaneously by the camera 5 of the apparatus. In addition there is shown the reflectivity 41 produced by replacing the wedge-shaped beam 100 with a collimated (i.e. non-converging) probe light beam which was simply mechanically scanned through the same range of angles of incidence. It can be immediately seen that when a wedge-shaped optical probe light beam is used, this generates strong resonance structures in the reflectivity spectrum on the form of a zero-order, then first-order, and then second-order optical resonances showing multiple dip-peak pairs 40 at successive angular (incidence) positions. This is not seen at all when a collimated probe light beam is used and is simply mechanically scanned across the same angles of incidence as was employed when using the wedge-shaped beam 100. Instead, only small dips in reflectivity are observed. These are principally due to scattering losses of probe light within the LW when the optical path of the light is increased greatly by the existence of optical resonance conditions within the LW. However, the absence of any curvature in the wavefront of the probe light means that a transverse modes do not exist in the probe light beam and so inter-mode coupling between transverse modes and longitudinal modes cannot occur. It is this inter-mode coupling which is believed to produce the strong dip-peak pair structures shown in Figure 4, as discussed above.

[86] The presence of the dip-peak pairs at the resonance angle was surprising, and a theoretical model was developed to explain the observation. The transfer matrix method was used to obtain the amplitude transmittance coefficients of the LW at different angles of incidence for TE and TM polarized light. This method is well known and its derivation will not be repeated here. It is based on the continuity conditions for the electric field across boundaries between adjacent layers that are derived from Maxwell's equations. From this, if the field at the beginning of a layer is known, then the field at the end of the layer can be derived by a matrix multiplication. A layer stack can be modelled using a system matrix, which is just the product of all of the layer matrices. The first and last layers are considered as semi-infinite. The reflection and transmission coefficients can be generated from the system matrix. Fresnel's approximation was then used to propagate the outgoing optical field to the detector. MATLAB was used to obtain the shifted FFT of the amplitude transmittance coefficient, which was multiplied with Equation 1.

H = elk7e-imi7u2 (1) where, k: wave vector (2-rr/A, m-1), z: distance to the detector (m), A: wavelength (m), u2: quadratic phase term.

[87] The absolute value of the inverse EFT of the result provided the output of the LW as seen by the detector. Theoretical reflectivity curves obtained using this approach for two LWs, where one comprised of a slab waveguide of refractive index of 1.34 (curves 51 and 52) and other of refractive index of 1.38 (curves 51 and 53), are provided in Figure 5. The thickness of the slab waveguides was such that only the zero-order mode was supported. The wavelength was 650 nm and the propagation distance was 5 cm. The dip-peak was only visible when the phase change was very sharp (5 -95% change over 0.175°), and not visible when the phase change was broader (5 -95% change over 1.505'). This shows that only very narrow waveguide resonances are visible by this method, which in turn requires a very low index waveguide.

[88] Figure 6(a) shows that the same effect occurs in single-and two-moded waveguides. This implies that this phenomenon can be observed in thick waveguides of low refractive index, but as the film increases in thickness, the higher order modes become progressively weaker. Curve 61 corresponds to a film thickness of 2 microns, and curve 62 corresponds to a film thickness of 4 microns. Figure 6(b) shows the evolution of the reflectivity changes with increasing propagation distance. Curves 63, 64, 63 and 66 correspond, respectively, to propagation distances of 5cm, 10cm, 20cm and 40cm. It can be seen that the peak-to-peak intensity change increases with the propagation distance.

[89] A LW was fabricated by spin coating 6% (w:v) solutions of poly(acrylamide) (poly(AAm)) at 4000 rpm and crosslinked via glutaraldehyde (GA). Blocks of poly(AAm) and copolymers are commonly prepared by three-dimensional radical polymerisation of monomers and a crosslinker, bis-acrylamide. The approach is, however, not well-suited to fabricate these thin films by spin coating because (1) the low viscosity monomer/ crosslinker solutions fly off without coating the substrate, (2) reducing oxygen inhibition of radical polymerisation in thin films is challenging, and (3) the hazardous nature of acrylamide/ bis-acrylamide. These limitations were overcome by crosslinking of linear polymers of poly(AAm) and poly(acrylamide-co-acrylic acid) (poly(AAm-AA) with GA in acidic conditions, and were used in this work to fabricate the LVVs.

[90] Chemicals and materials [91] In an example, 1 mm thick standard microscope glass slides were purchased from VVVR (Leicestershire, UK). Ethanol, toluene, (3-Aminopropyl)triethoxysilane (APTES), 1 M hydrochloric acid (HCI), poly(acrylamide) (poly(AAm) with Mn: 150k), poly(acrylamide-co-acrylic acid) partial sodium salt (poly(AAm-AA), Mw: 520k, Mn: 150k, acrylamide -80%), 25% (v:v) glutaraldehyde (GA), poly(ethylene glycols) (PEGs) of different molecular weights, (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) (HEPES), 1 M sodium hydroxide (NaOH), reactive blue 4 (RB4), bovine serum albumin (BSA, A2153), biotin Anti-IgG (B3773) and IgG (15131) were bought from Sigma-Aldrich (Gillingham, UK). Decon 90, Glycerol (Mw: 92) and N(3-Dimethylaminopropy1)-1V-ethylcarbodiimide hydrochloride (EDC) were purchased from Fisher (Loughborough, UK). Streptavidin (N7021S, contained 140 mM, 10 mM KCI, NaCI, 8 mM sodium phosphate and 2 mM potassium phosphate) and N-Hydro>cysulfosuccinimide sodium salt (sulfo-NHS) were supplied by New England Biolabs (Hertfordshire, UK) and Cambridge Biosciences (Cambridge, UK) respectively.

[92] Wave guide fabrication [93] Glass slides were cut into squares of -25.4 mm by 25.4 mm using a diamond scribe and cleaned in Decon 90 solution, water and ethanol for 30 mins after each step in an ultrasonic bath (Ultrawave U300H). The dried slides were treated with 1% (v:v) APTES in toluene for 30 min, washed with toluene, dried and immediately used.

[94] The required amount of either poly(AAm) or poly(AAm-AA) was dissolved in 1 M HCI and appropriate volume of GA was added to the polymer solution before spin coating. The concentration of GA in the polymer solution was 250 ppm. -80 pl of the polymer solution was dispensed on a glass slide and spin speed was varied between 2000 rpm and 4000 rpm. The solution was spun for 30 s at an acceleration of 200 rpm/ s and the films were allowed to crosslink overnight.

[95] Wave guide characterisation [96] The porosity of the waveguides to glycerol and PEGs of different molecular weights prepared in 100 mM HEPES buffer, pH 7.4 was tested by recording the resonance angle as glycerol/ PEG solutions were pumped through the flowcell placed on the top of the waveguide. The concentration of glycerol solutions was between 0.125% (v:v) and 2% (v:v). The concentration of solutions of PEGs of different molecular weights was fixed at -1% (w:v). A buffer wash was performed between PEG solutions.

[97] To study the biosensing capability of the waveguide, the poly(AAm-AA) film was activated in-situ by reacting the carboxylic acid groups with 1.8 mg of EDC and 2 mg of sulfo-NHS dissolved in 100 mM HEPES, pH 5.8 buffer re-circulated through the flowcell for -1 h. The film was washed with HEPES buffers of pH 5.8 and then pH 7.4. The remaining solutions were prepared in 100 mM HEPES, pH 7.4. -0.2 mg/ ml of streptavidin and BSA solutions were pumped through one channel each of the flowcell to react the amine groups in the proteins with the EDC-sulfo NHS activated poly(AAm-AA) film. The regions of the waveguide positioned under the flow channels through which streptavidin and BSA solutions were pumped provided the measurement and reference signals respectively. -67.5 nM biotin Anti-19G and -133.3 nM IgG solutions were pumped through both the flow channels sequentially. A buffer wash was performed in between the protein solutions and resonance angle was monitored as different solutions were introduced on the waveguide.

[98] Instrumentation [99] Referring to Figure 1 again, a BK7 equilateral prism 3 (Qioptic Photonics, Denbighshire, UK) was used to couple light in and out of the hydrogel waveguide 2. The light emitter LED 10 (TL-6, iC-Haus, Bodenheim, Germany) and 6.6 Mpixel CMOS camera 5 (PL-B781, Pixelink, Ottowa, Canada) were mounted on rails connected to goniometers to allow radial and angular freedom respectively. The output of the LED 10 was collimated, TE-polarised and then passed through a 40 mm focal length cylindrical lens 6 and an aperture 7 to form a wedge beam 100 to probe the hydrogel waveguide with a range of angles of incidence simultaneously. The camera allowed a 7.7 mm wide section of the LW to be imaged, which allowed both the flow channels to be captured in a single frame.

[100] The collimated mechanical scan (trace 41 in Figure 4) was obtained using a TE-polarised laser as the light source (Acculase, RS Components, Northamptonshire, UK) with a peak wavelength of 650 nm and a power of 5 mW and photodiode (OSD100-6, Centronic, Surrey, UK). The LED light source and lens train (items 6-10 of Fig.1) were not used in that case.

[101] Fluids were pumped through the flow cell 4 using a peristaltic pump (Minipuls® 3, Gilson, Bedfordshire, UK) at a flow rate of 0.2 ml min-1. The flowcell 4 was CNC machined from 3 mm thick black PMMA forming two recessed cavities to obtain 2 mm wide and 0.2 mm deep channels surrounded by grooves 1 mm wide and 0.75 mm deep in which 0-rings were mounted. The flow cell 4 was placed on the waveguide 2 and held in place using a water-cooled fixture maintained at 20 °C.

[102] The refractive index of the solutions of sample material flowing through the flow cell 4 was measured using RFM900-T refractometer (Bellingham and Stanley, Kent, UK) with an accuracy of ±1 x1 0-5.

[103] Refractive index sensitivity (RIS) [104] To obtain single-moded LWs, the concentration of poly(AAm) solution used to fabricate the films was reduced to 3% (w:v) and the output of the resulting waveguide is provided in the inset in Figure 7.

[105] The angular position of the dip and peak of each dip-peak pair was found in real-time using a centre-of-gravity based algorithm. Subsequently, the average of the angular positions of the dip and peak of the dip-peak pair was recorded as different concentrations of glycerol solutions were introduced on the LW (curve 70 of Figure 7). Figure 4 shows that the angular positions of the modes as determined by a mechanical scan 41 corresponds closely to the average position of the dip-peak pair 40. For sensing purposes, only changes in the dip-peak angular position are of interest rather than their absolute value. Averaging the dip and peak positions resulted in a small improvement in signal-to-noise ratio. The intensity of the dip and peak did not change significantly as the glycerol concentration changed, because the refractive index contrast between the waveguide and sample remained substantially constant. The slope of the calibration plot of the angular positions of the dip and peak of the dip-peak pair versus refractive index of different concentrations of glycerol solutions provided the RIS of the LW for glycerol. The RIS of the LW with poly(AAm) waveguide to glycerol was 119.58 ± 0.37 and 119.42 ± 0.63 ° RIU-1 at the 95% confidence level for channels 1 and 2 respectively. A t-test on the RIS values indicates that there is no significant difference in the sensitivity of the two channels at the 95% confidence level, so the average of the two values (119.5 ° RIU-1) is used in subsequent calculations. The minimum detectable change in refractive index, based on three times the standard deviation of the angular noise, was 4.5 x 10-6 RIU.

[106] The RIS is a key figure of merit for comparing the performance of label-free optical biosensors. The RIS of the LW to glycerol is comparable to SPR on gold, which has a RIS between 95 and 204° RIU-1 depending on the prism index and configuration. In SPR, the gold film is typically deposited on a high refractive index substrate such as SF10, which is several times more expensive than the soda-lime glass substrate used to fabricate the LWs reported in this work (soda-lime microscope slides of dimensions 25 mm by 75 mm). Well-controlled and high quality gold films required for SPR are deposited using vacuum evaporators. In contrast, the poly(AAm) waveguides in the LWs were deposited via spin coating. Solution-based fabrication methods such as spin coating are well suited for cost-effective large-scale manufacturing. The peak-dip pairs in the reflectivity curves of the LWs were observed using low-cost LEDs and read out was performed using a simple CMOS camera. A combination of the use of affordable materials, fabrication process and instrumentation along with the RIS comparable to SPR contributes towards the suitability of the developed LW for point-of-care (PoC) diagnostics.

[107] To obtain high sensitivity to biochemical species such as proteins, the waveguide should be porous enough to allow these to diffuse in. Thus, being able to detect large molecular weight analytes with improved sensitivity requires high RIS for macromolecules. The porosity was determined by using PEG solutions of similar refractive index but different molecular masses, This showed that the shift in resonance angle corresponding to 1% (w:v) PEGs of M",,, 10 kD, 100 kD and 300 kD solutions is -41%, -29% and -21% of the value observed for glycerol solution of similar refractive index (curve 71 of Figure 7). Since the evanescent field sensitivity is between 16% and 24% of the total RIS, this shows that the 10 and 100 kD PEGs were able to diffuse into the hydrogel layer to some extent, thereby interacting with evanescent field and significant proportion of the optical mode confined in the waveguide. These RIS values for PEGs are significantly higher than the ones obtained with LWs made of agarose, where PEGs were largely excluded from penetrating the waveguide. Many biomolecules of clinical significance and ligands such as antibodies have molecular weight in the range of 50 -200 kD.

[108] Figure 7, the inset, also shows that the variation in the response of the LW for different glycerol and PEG solutions across both 2 mm wide flow channels was <1.55%. Minimal variations in the area-to-area response is required so that the shift in the resonance angle because of common-mode effects (e.g. temperature, composition of sample matrix) and analyte-ligand interactions can be de-convoluted using spatially separated sensor and reference regions of the sensor.

[109] In chlorine, pH and bio-sensing applications, it is most preferable that the hydrogel waveguide layer is porous to allow the analyte(s) (e.g. hydrogen ions or chlorine) to diffuse into the waveguide. Multi-analyte sensing may be carried out by spatially patterning the sensing molecules over the surface of the sensor. The wedge beam may provide spatial resolution to allow different areas of the chip to be monitored. Real-time monitoring of analyte binding can be carried out, e.g. for the antibody-antigen binding.

[110] pH-sensing [111] A suitable non-ionisable cross-linked polymer backbone may be used to form a hydrogel matrix of the waveguide. An example is polyacrylamide. This material may be post-polymerisation modified with pendant ionisable groups (such as amines or carboxylic acids) that cause the polymer to swell/shrink as the pH changes and the ionisable groups change their charge. For example, amines will be protonated at pH below their pKa, and the positive charges on the protonated amines will cause the polymer to swell at pH below the pKa.

[112] In another example, the hydrogel waveguide may incorporate monomers with ionisable groups in the cross-linked polymer backbone used to form a hydrogel matrix of the waveguide thereby to form an ionisable copolymer. Examples include acrylamide-acrylic acid copolymers. By changing the percentage of ionisable groups in the polymer, the pKa can be modulate to change the pH range over which the sensor responds.

[113] Chlorine sensing [114] The hydrogel waveguide matrix may comprise a suitable non-reactive cross-linked polymer backbone which has attached to it reversible chlorine-reactive groups. An example is cyanuric acid, where chlorine can reversibly displace hydrogen atoms on the triazine nitrogens, attached to the polymer through one of the nitrogen atoms in the triazine ring.

[115] Bio-sensing [116] Biosensing may be performed this using antibody-antigen binding at the matrix of the hydrogel waveguide. Generally, this method may be applied to any sensing mechanism that involves a change in refractive index. Examples include aptamers, DNA, enzymes, receptors.

[117] In an example, bio-sensing is demonstrated herein using LWs made of poly(AAm-AA) because it provides functional groups that allowed covalent attachment of analyte-specific moieties (e.g. antibodies) to the hydrogel waveguide. 3% (w:v) polymer solution containing 250 ppm GA was spin coated at 2000 rpm. The films were allowed to crosslink overnight following which, the carboxylic acid groups in the regions of the waveguide positioned under the sensor and reference flow channels were activated with EDC-sulfo NHS to allow covalent attachment of streptavidin and BSA respectively. Subsequently, the same solutions were pumped through both sensor and reference channels, and their corresponding response to each solution was recorded. Figure 8(a) shows absolute shifts in the resonance angle of the sensor and reference regions. Figure 8(b), on the other hand, provides the differential response (curve 86), which was obtained by subtracting the average response of the reference from the sensor region. In particular, figure 8 shows a further diagram illustrating changes in the detected angular position of an angle at which probe light out-coupled from a leaky waveguide achieves a peak or a trough in light intensity, according to an example of the invention. In Figure 8(a), the results for three different samples (curves 80, 81 and 82) are compared to the results for three different reference channels (curves 83, 84 and 85).

[118] The shift in the resonance angle of the sensor and reference regions because of BSA was very similar and hence the differential response was negligible. Additionally, the shift in the resonance angle of the sensor and reference regions because of BSA decreased with a buffer wash. This implies that BSA was either weakly bound or unbound to ligands in the sensor and reference regions, and the shift in the resonance angle because of BSA was largely because of the refractive index different between the protein solution and buffer. A downward drift in the baseline was observed in the absolute response of the sensor and reference regions, but the baseline of the sensorgram obtained by taking a difference of the two was flat. In contrast, when biotin Anti-IgG was introduced, the response in the sensor channel was much higher than the reference channel resulting in a net shift of 0.014° in the differential trace.

[119] As biotin Anti-IgG was strongly bound to streptavidin, no change in the resonance angle was observed because of the buffer wash. Similarly, IgG bound selectively to biotin Anti-IgG in the sensor channel. The binding affinity between IgG and Anti-IgG is lower than between streptavidin and biotin. Thus, the rate of change of resonance angle in Figure 8(b) because of IgG-Anti-IgG binding was lower than streptavidin-biotin.The invention exploits the unexpected discovery by the inventors, that leaky modes of very low index waveguides can be seen in the reflectivity curve as dip-peak pairs without any additional means of visualisation such as metal layers or dyes. This phenomenon was only observed when the waveguide was illuminated through a prism with a converging wedge beam produced by a cylindrical lens. The similarity of these dip-peak pairs in reflectivity to the diffraction patterns observed when light propagates past opaque edges or phase steps led us to hypothesise, which was subsequently proved by theoretical modelling, that these patterns arose via a similar mechanism.

[120] As the effective index of the waveguide layer changes, the angular position of this dip-peak pair changes, allowing binding events to be monitored in real-time. Relying solely on diffraction to visualize the waveguide resonances has resulted in a very simple optical biosensor, consisting of a glass slide with polyacrylamide and copolymers waveguide layer. The LW has a refractive index sensitivity of -119.5 ° RIU-1, which is comparable to the current market-leading technology, surface plasmon resonance (SPR). The minimum detectable change in refractive index of the LWwas 4.5 x 10-6 RIU. Biosensing was demonstrated using a poly(AAm-AA) waveguide by activating the carboxylic acid groups with EDC/NHS to allow covalent immobilisation of streptavidin, followed by non-covalent binding of biotin Anti-IgG and then IgG.

[121] This invention advances the field of label-free optical sensors because the reported LW may be fabricated using scalable manufacturing processes, is made of widely available low cost materials, requires affordable instrumentation, and has comparable performance to SPR. This in turn makes the LW biosensor highly attractive for applications such as point-of-care (PoC) diagnostics, or for pH or Chlorine sensing.

[122] The waveguide may simply be a thin (1 -5 microns) porous hydrogel film 2 on a glass or plastic substrate 2A (e.g. Fig.3A). The glass or plastic substrate may be are coupled to a prism 3 using index matching fluid or gel (not shown), while polymer substrates can have the prism moulded in to avoid using index matching fluids or gels.

Claims (25)

1. CLAIMS1. A sensor for sensing a change in a refractive index caused by the presence of a substance, comprising: an optically transparent substrate; an optical waveguide formed upon a surface of the substrate thereat to form a first optical interface with the surface of the substrate for receiving light therethrough, wherein the optical waveguide comprises an external surface opposing the first optical interface for forming a second optical interface by contact with said substance; a light source configured to inject into the optical waveguide a convergent beam of probe light having a curved optical wavefront such that the probe light is guided along the optical waveguide between the first and second optical interfaces; a detector configured to receive probe light retumed from the waveguide via the first optical interface across an interval of oblique angles relative to the first optical interface; wherein the detector is arranged to detect a change in an angle within said interval at which the received probe light achieves a peak or a trough in light intensity, and to determine a change in the refractive index at the second optical interface caused by the presence of the substance according to the change in said angle.
2. 2. A sensor according to any preceding claim in which the optical waveguide is formed from a material having a refractive index not exceeding 1.38000 in respect of light having a wavelength of 650nm
3. 3 A sensor according to any preceding claim in which the optical waveguide is formed from a material having a refractive index not exceeding 1.36800 in respect of light having a wavelength of 650nm
4. 4. A sensor according to any preceding claim in which the optical waveguide is formed from a material having a refractive index on the range from 1.30000 to 1.36800 in respect of light having a wavelength of 650nm.
5. 5. A sensor according to any preceding claim in which the optical waveguide comprises a material forming a hydrogel matrix.
6. 6. A sensor according to preceding claim in which the optical waveguide is formed from a hydrogel or a dehydrated hydrogel matrix which is re-hydratable to form a hydrogel.
7. 7. A sensor according to any preceding claim in which the light source is arranged to form said convergent beam of probe light as a wedge-shaped beam which is convergent in a first direction orthogonal to the axis of the beam of probe light, but which is not convergent in a second direction which is orthogonal to both the first direction and said axis.
8. 8. A sensor according to any preceding claim in which the optical waveguide is a planar waveguide in which the first optical interface and the second optical interface are disposed in substantially plane-parallel relative opposition and separated by a substantially uniform waveguide thickness of at least 1pm.
9. 9. A sensor according to claim 8 in which said waveguide thickness does not exceeding 10pm.
10. 10. A sensor according to any preceding claim in which the detector is arranged to determine: a first said angle at which the received probe light achieves a trough in light intensity; a second said angle at which the received probe light achieves a peak in light intensity; a mean angle as an average of the first angle and the second angle; a change in the refractive index at the second optical interface according to a change in said mean angle.
11. 11. A sensor according to any preceding claim in which the second said angle is greater than the first said angle.
12. 12. A sensor according to any preceding claim in which said peak is a maximum in said light intensity within said interval.
13. 13. A sensor according to any preceding claim in which said trough is a minimum in said light intensity within said interval.
14. 14. A sensor according to claims 12 and 13 in which said peak is the first peak that follows said trough in angular distance along said interval.
15. 15. A sensor according to any preceding claim in which the optical waveguide is a leaky waveguide configured to support leaky optical modes in respect of the injected probe light.
16. 16. A sensor according to any preceding claim in which the optical waveguide comprises a porous material adapted to admit diffusion of a said substance into the optical waveguide via the second optical interface.
17. 17. A pH sensor comprising a sensor according to any preceding claim in which the optical waveguide comprises a material forming a hydrogel matrix adapted to swell or shrink thereby to change the refractive index of the optical waveguide in reaction to changes in the pH of said substance at said second optical interface.
18. 18. A chlorine sensor comprising a sensor according to any preceding claim in which the optical waveguide comprises a material forming a hydrogel matrix containing chlorine-reactive groups adapted change the refractive index of the optical waveguide in reaction to chlorine at said second optical interface.
19. 19. A biosensor comprising a sensor according to any preceding claim in which the optical waveguide comprises a material forming a hydrogel matrix adapted to permit the binding thereto of biological analyte materials via said second optical interface.
20. 20. A method for sensing a change in a refractive index caused by the presence of a substance, comprising: providing an optically transparent substrate bearing an optical waveguide upon a surface of the substrate defining a first optical interface with the surface of the substrate for receiving light therethrough; by contact with said substance, forming a second optical interface at an external surface of the optical waveguide opposing the first optical interface; injecting into the optical waveguide a convergent beam of probe light having a curved optical wavefront and guiding the probe light along the optical waveguide between the first and second optical interfaces; receiving probe light returned from the waveguide via the first optical interface concurrently across an interval of oblique angles relative to the first optical interface; detecting a change in an angle within said interval at which the received probe light achieves a peak or a trough in light intensity, and determining a change in the refractive index at the second optical interface caused by the presence of the substance according to the change in said angle.
21. 21. A method according to claim 20 including forming said convergent beam of probe light as a wedge-shaped beam which is convergent in a first direction orthogonal to the axis of the beam of probe light, but which is not convergent in a second direction which is orthogonal to both the first direction and said axis.
22. 22. A method to any of claims 20 and 21 including determining: a first said angle at which the received probe light achieves a trough in light intensity; a second said angle at which the received probe light achieves a peak in light intensity; a mean angle as an average of the first angle and the second angle; and a change in the refractive index at the second optical interface according to a change in said mean angle.
23. 23. A method for pH sensing comprising the method according to any of claims 20 to 22 in which the optical waveguide comprises a material forming a hydrogel matrix adapted to swell or shrink thereby to change the refractive index of the optical waveguide in reaction to changes in the pH of said substance at said second optical interface.
24. 24. A method sensing chorine comprising the method according to any of claims 20 to 22 in which the optical waveguide comprises a material forming a hydrogel matrix containing chlorine-reactive groups adapted change the refractive index of the optical waveguide in reaction to chlorine at said second optical interface.
25. 25. A method for sensing a biological analyte material comprising the method according to any of claims 20 to 22 in which the optical waveguide comprises a material forming a hydrogel matrix adapted to permit the binding thereto of biological analyte materials via said second optical interface.