WO2012076709A1

WO2012076709A1 - Magnetic acoustic resonator sensor

Info

Publication number: WO2012076709A1
Application number: PCT/EP2011/072378
Authority: WO
Inventors: Chris Lowe; Adrian Stevenson; Bernardita Araya-Kleinsteuber
Original assignee: Cambridge Enterprise Limited
Priority date: 2010-12-10
Filing date: 2011-12-09
Publication date: 2012-06-14
Also published as: CN105277612A; CN105259249A; CN102918389A; CN102918389B; GB201020939D0; CN105259249B; CN105277612B

Abstract

A sensing apparatus comprising: a resonator, a sensor and a detector, the sensor being mechanically coupled to the resonator, the sensor comprising a sensor material which changes between a first state and a second state when exposed to a change in the surrounding environment, wherein the sensor is driven by the resonator and the detector is responsive to the change in state of the sensor material, wherein the sensor material is in the form of an acoustically thick layer.

Description

MAGNETIC ACOUSTIC RESONATOR SENSOR

FIELD OF THE INVENTION

The present invention relates to a sensing apparatus and preferably a chemically activated electro-mechanical sensing apparatus using hydrogels.

BACKGROUND OF THE INVENTION

Hydrogels that are able to change shape or volume when exposed to different chemical environments have been used as pH, ion, chemical, gas and temperature sensors and as mechanical actuating elements. See, for example, Bashir, R., et al., Micromechanical cantilever as an ultrasensitive pH microsensor, Applied Physics Letters, 2002. 81(16): p. 3091-3093 and Zhang, L. and W.R. Seitz, A pH sensor based on force generated by pH- dependent polymer swelling, Analytical and Bioanalytical Chemistry, 2002. 373(7): p. 555- 559, van der Linden, H., et al., Development of stimulus-sensitive hydrogels suitable for actuators and sensors in microanalytical devices. Sensors and Materials, 2002. 14(3): p. 129- 139, Mayes, A.G., et al., Metal ion-sensitive holographic sensors. Analytical Chemistry,

2002. 74(15): p. 3649-3657, Mayes, A.G., et al., A holographic alcohol sensor. Analytical Chemistry, 1999. 71(16): p. 3390-3396, Mayes, A.G., et al., A holographic sensor based on a rationally designed synthetic polymer. Journal of Molecular Recognition, 1998. 11(1-6): p. 168-174, Millington, R.B., et al., A hologram biosensor for proteases. Sensors and Actuators B-Chemical, 1996. 33(1-3): p. 55-59, Blyth, J., et al., Holographic sensor for water in solvents. Analytical Chemistry, 1996. 68(7): p. 1089-1094, Herber, S., W. Olthuis, and P. Bergveld, A swelling hydrogel-based P-C02 sensor. Sensors and Actuators B-Chemical,

2003. 91(1-3): p. 378-382, Kuckling, D., et al, Photo cross-linkable poly(N- isopropylacrylamide) copolymers III: micro-fabricated temperature responsive hydrogels. Polymer, 2003. 44(16): p. 4455-4462 and Hilt, J.Z., et al., Ultrasensitive biomems sensors based on microcantilevers patterned with environmentally responsive hydrogels. Biomedical Microdevices, 2003. 5(3): p. 177-184. Optical detection of the dimensional changes of either hydrogel crystals or diffraction gratings has provided a convenient sensorimetric signal based on a colour change of the modified hydrogel, the latter providing a more generic detection platform. See, for example, Asher, S.A., et al., Photonic crystal carbohydrate sensors: Low ionic strength sugar sensing. Journal of the American Chemical Society, 2003. 125(11): p. 3322-3329 and Marshall, A.J., et al., PH-sensitive holographic sensors. Analytical Chemistry, 2003. 75(17): p. 4423-4431.

However, there are several aspects of hydrogels to be considered for specific applications. See, for example, Hoffman, A.S., Hydrogels for biomedical applications. Advanced Drug Delivery Reviews, 2002. 43: p. 1-12. One of the key considerations for optical hydrogel systems is the maintenance of the position of particles that define the subsequent colour change. This requires that a hydrogel must retain sufficient rigidity for stability, but at the same time provide low cohesive forces to allow measurable volumetric changes to occur. These mechanical aspects of hydrogels have been investigated by straining rectangular elements of poly-HEMA (Hydroxyethyl methacrylate) several centimetres in size in order to measure their elasticity at different pH values. See, for example, Johnson, B., et al. Mechanical properties of a pH sensitive hydrogel in SEM Annual conference proceedings. 2002. Milwaukee, WI.

An alternative method of sensing uses acoustic wave sensors which respond to viscoelastic solutions. An acoustic wave sensor comprises a resonator which is stimulated to vibrate in the viscoelastic solution. A change in the viscosity of the solution (e.g. due to a chemical reaction) causes a change in the frequency/amplitude of vibration of the resonator which can be detected.

An adaptation of the aforementioned sensor comprises providing a mono-molecular film on the resonator, the mono-molecular film acting as a probe for the surrounding chemical/physical environment. The mono-molecular film can react with molecules in the surrounding solution for example resulting in a change in mass/volume of the resonator which changes the frequency/amplitude of vibration of the resonator. Such an arrangement has led to the detection of nucleotides using a mono-layer of hydrogel. See, for example, Kanekiyo, Y., et al., Novel nucleotide-responsive hydrogels designed from copolymers of boronic acid and cationic units and their applications as a QCM resonator system to nucleotide sensing. Journal of Polymer Science Part a-Polymer Chemistry, 2000. 38(8): p. 1302-1310, Tang, A.X.J. , et al., Immunosensor for okadaic acid using quartz crystal microbalance, Analytica Chimica Acta, 2002. 471(1): p. 33-40, Nakano, Y., Y. Seida, and K. Kawabe, Detection of multiple phases in ecosensitive polymer hydrogel, Kobunshi Ronbunshu, 1998. 55(12): p. 791-795, and Serizawa, T., et al., Thermoresponsive ultrathin hydrogels prepared by sequential chemical reactions. Macromolecules, 2002. 35(6): p. 2184- 2189.

Acoustic sensors can thus behave as a probe for the surrounding environment without the need to incorporate diffraction gratings. However, in the above described arrangements, perturbations of typically < 0.01% in the acoustic frequency occur which require careful tracking in order to detect the change. That is, current acoustic wave sensors suffer from the problem that the effects in the sensor occurring due to a change in the physical/chemical environment are small and require very sensitive measurement in order to obtain useful information about the change.

SUMMARY OF THE INVENTION

The present invention seeks to utilize changes in the physical and/or chemical properties of a driven sensor to monitor changes in the surrounding environment. In particular, the present invention seeks to solve the problem of low sensitivity of current acoustic wave sensors outlined above.

According to a first aspect of the present invention there is provided a sensing apparatus comprising: a resonator, a sensor and a detector, the sensor being mechanically coupled to the resonator, the sensor comprising a sensor material which changes between a first state and a second state when exposed to a change in the surrounding environment, wherein the sensor is driven by the resonator and the detector is responsive to the change in state of the sensor material, wherein the sensor material is in the form of an acoustically thick layer. A principal advantage of the present acoustic approach is that it is possible to recover information about the internal chemical forces arising from cross-linking and copolymerisation, which are evident from hydrogel elasticity, but not apparent from any dimensional changes collected optically. An advantage of embodiments of the present invention is that the acoustic phase changes across the thickness of the hydrogel film (acoustically thick film) are large and close to Pi/2 so as to substantially modify resonance to the point of extinction and hence provide a useful mechanical switching action between states of vibration and no vibration.

Accordingly, an advantageous technical effect that embodiments of the present invention have over the prior art is that the shift in amplitude of the acoustic signal of the resonator when a change in the chemical/physical environment occurs is large and thus easier to detect when compared with prior art arrangements in which there is only a small shift in the acoustic frequency. The sensor layer, which may be a biological recognition layer, itself becomes a resonant structure that stores or dissipates energy. The sensor response then becomes dominated by the chemical composition of the sensor layer or over-layer. A hydrogel film is one successful example of such an over-layer.

This advantageous technical effect is achieved by providing a sensor material in the form of an acoustically thick layer as distinct from the thin layers disclosed in the prior art. That is, the provision of an acoustically thick film leads directly to the more extreme deviation in the acoustic signal.

For viscoelastic materials, the transition between acoustically thin behaviour and acoustically thick behaviour is most conveniently defined by the phase factor. Accordingly, acoustically thick films have a thickness that approximately satisfies the following equation:

ω

where ω is the angular frequency, p is the density, G is the shear modulus, and t is the film thickness.

An acoustically thick layer is one which can support a significant phase shift in the acoustic wave as it travels perpendicular to the layer. This limitation discounts the monolayers disclosed in the prior art arrangements. Such monolayers cannot support an acoustic phase shift perpendicular to the layer. Furthermore, such monolayers cannot support an acoustic phase shift parallel to the layer unless high frequencies according to the above equation are used.

In the context of the present application "acoustic vibrations" refer to standing acoustic waves reflected between upper and lower boundaries of the sensor.

Preferably, the detector comprises an electromagnetic field generator capable of being arranged to direct an electromagnetic field towards the sensor. This is advantageous as the sensor can be used remotely without the need for a connecting wiring. Preferably, the electromagnetic field generator and the detector comprise a common structural element for generating an electromagnetic field and detecting an electromagnetic field. This provides a compact arrangement. Advantageously, the electromagnetic field generator is tuneable. This gives the apparatus a greater variability and sensitivity in use. The electromagnetic field generator may be a spiral coil and the sensing apparatus may comprise a signal generator and a lock-in amplifier connected to the electromagnetic field generator and the detector to provide greater sensitivity. The detector may comprise a differential diode demodulation circuit for subtracting a detected signal from a signal produced by the signal generator. In one arrangement the resonator comprises a magnet arrangeable to direct a magnetic field towards the sensor. This arrangement provides a magnetic acoustic resonant sensor. The resonator may comprise an exciter mechanically coupled to the sensor which may be a piezoelectric material such as a layer of quartz. The piezoelectric layer is preferably 50 to 1000 μιη thick. Other sensing apparatus can be used to measure the detector's electrical impedance and hence the acoustic response. For example, a commercial impedance analyser or other custom made impedance measuring circuit may be incorporated.

The exciter could be derived from a magnetostrictive material, where the magnetic component of the resonator's electromagnetic field is used to excite the acoustic waves. The exciter may comprise a metallic material. This may be in the form of a layer which is preferably 50 to 1000 μηι thick.

The sensor layer, comprising the acoustically thick layer of sensor material, is preferably 0.1 μηι to 1 mm thick, more preferably 0.1 μηι to 100 μηι thick, more preferably still 0.1 μιη to 10 μηι thick and most preferably 0.5 μηι to 5 μιη thick. Advantageously, the sensor layer is more than one molecule thick.

The sensor material is preferably a hydrogel, for example hydroxyethyl methacrylate-co- methacrylate poly(acrylamide-co-3-acrylamido phenyl boronic acid) or poly(acrylamide-co- 2-acrylamido phenyl boronic acid).

Preferably, the sensing apparatus is a chemically actuated electro-mechanical sensing apparatus. The sensing apparatus may be used in a method of sensing. Furthermore, the sensing apparatus may be used as a switch or in a method of continuous monitoring. Additionally, the sensing apparatus may be used in a method of controlling a system based upon a change in the surrounding environment.

According to another aspect of the present invention there is provided a method of making a composite probe comprising a resonator element which may be piezoelectric and an acoustically thick hydrogel film thereon, the method comprising: preparing a monomer mixture for the hydrogel; applying the monomer mixture to a release layer such as an aluminised polyester sheet; applying the resonator element to the monomer mixture; polymerising the monomer mixture to form the acoustically thick hydrogel film adhered to the resonator element thereby forming the composite probe; and peeling the composite probe from the release layer. The resonator element may be treated with a binding agent such as trimethoxysilylpropyl methacrylate prior to application of the monomer mixture to promote binding. In one embodiment, the hydrogel is prepared from a mixture of 2-hydroxyethyl methacrylate (HEMA), ethylene dimethyacrylate (EDMA), methacrylate (MAA) and a photoinitiator such as dimethoxyphenylacetone. In another arrangement, the hydrogel is prepared from a mixture of 3-acrylamido phenyl boronic acid, acrylamide and N,N- methylenebis acrylamide. In a further arrangement, the hydrogel is prepared from a mixture of 2-acrylamido phenyl boronic acid, acrylamide and Ν,Ν-methylenebis acrylamide.

Embodiments of the present invention seek to utilize changes in acoustic properties of materials brought on by changes in their physical and/or chemical environment in order to provide a chemically activated electro-mechanical sensing apparatus.

In one embodiment, a hydrogel is responsive to change in chemical environment, preferably wherein the change in chemical environment is a change in pH. In this embodiment, the hydrogel is typically a polymer which is responsive to pH by changes in its viscoelastic behaviour. The volume of hydrogel may also change in response to changes in pH. A hydrogel material suitable for this purpose is hydroxyethyl methacrylate-co-methacrylate, which is a co-polymer of 2-hydroxyethyl methacrylate and methacrylate, preferably cross- linked with an amount of ethylene di-methacrylate (also known as ethylene-glycol- dimethacrylate). The amounts of methacrylate used to form the hydrogel are usually around 6 mol% with an amount of ethylene dimethacrylate typically in the range 1.5 to 7.5 mol %. Generally, hydrogels with more than 3.5 mol % ethylene dimethacrylate tend to become more rigid and less flexible. The balance of hydroxyethyl methylacrylate is in the range 86.5 mol % to 92.5 mo 1%.

Unlike conventional quarz crystal microbalances the apparatus of the present invention can operate wirelessly thereby eliminating non-disk electrodes. As compared with conventional operating frequencies of 5 MHz, this allows for an extension of the operating frequency to a range from 6 MHz to 1.1 GHz. In addition, wireless operation enables the resonator and sensor part of the sensing apparatus to be implantable, for example into a human or animal subject. Here, the detector may be operated ex vivo, remote from the resonator and sensor. In another embodiment, the hydrogel is responsible to change in chemical environment wherein the change in chemical environment is a change in concentration of an analyte. The analyte may be a physiological analyte such as a sugar. Such sugars may include glucose, fructose, galactose and mannose. Glucose is a particularly important physiological analyte, having a key involvement in disease conditions such as diabetes mellitus as well as in fermentation processes and the metabolism in living cells.

In one arrangement, the hydrogel is a polymer containing phenyl boronic acid groups which are typically pendant groups on a polymer main chain. Preferably, the polymer is poly(acrylamide-co-3-acrylamido phenyl boronic acid). This polymer is capable of binding glucose and other sugars and it is thought that the boronic acid moieties of the polymer bind with the cw-diols of the sugars.

In one arrangement, the sensing apparatus of the invention comprises a resonator and sensor which are in the form of an implantable device for implanting into a human or animal subject. In this arrangement, the detector is responsive to the sensor ex vivo thereby enabling remote monitoring of a physiological analyte such as glucose in a simple and non-invasive way. Accordingly, an implantable device is provided for implanting into a human or animal subject and comprises a resonator and a sensor as described herein in which the sensor is mechanically coupled to the resonator. The detector may be used with the implantable device in a method of remote sensing of a physiological state of the subject, wherein the detector is responsive to the sensor material of the sensor. The sensing of the physiological state of the subject may comprise sensing of an analyte, which sensing may be continuous monitoring of analyte concentration, for example.

Unlike enzyme-based commercial glucose sensors and monitors which tend to have poor stability in long term applications and may be difficult to sterilise for in vivo use, the sensing apparatus of the present invention is robust and readily implantable into a suitable subject.

In embodiments of the presently proposed apparatus and method the signal shift of the amplitude is extreme, and may be 95% or more. As a consequence it is easier to detect, simplifying the sensor instrument considerably. That is, the very large signal changes observed can lead to much simpler and less costly instrumentation. For example, the lock-in can be removed and conventional less sensitive electrical impedance measuring approaches adopted. Furthermore, embodiments of the present invention are more sensitive to physical and chemical changes than prior art arrangements.

For a better understanding of the present invention and to show how the same may be carried into effect reference will now be made by way of example to the accompanying drawings in which:

Figure 1(a) shows a hydrogel switch exposed to an electromagnetic field generated and monitored by a measurement system incorporating a signal generator, a detector and a lock-in amplifier connected to a PC (not shown) running Lab VIEW™ - a contacting solution is retained on an upper surface of the hydrogel switch with an o-ring and plastic cell (not shown);

Figure 1(b) shows a fundamental resonance peak of a quartz disc compared to the same disc after coating with a 10 μηι hydrogel film of poly-HEMA - the top surface of both the disc and hydrogel composite are exposed to solutions of pH 3.5;

Figure 2 shows a comparison between the shear wave resonance spectra of three separate hydrogel-quartz composites showing significant attenuation at 100MHz;

Figure 3(a) shows a phase contrast image of the edge of a hydrogel film indicating its thickness to be approximately 3 μιη;

Figure 3(b) shows the frequency shift at the fundamental and the third harmonic, after coating a quartz transducer with a hydrogel layer prepared originally with 5 (disc 1), 8 (disc 2), 12 (disc 3), and 15 (disc 4) μΐ of monomer mix;

Figure 4 shows the acoustic resonance harmonics of a single hydrogel-quartz composite following exposure to pH 3.5 and pH 8 - expansion of the film leads to a reduction in frequency at 6.7MHz, cessation of the resonance at 33MHz and an increase in frequency at 60MHz;

Figure 5(a) shows the switching action of a hydrogel-quartz composite with a more uniformly thick hydrogel film - pH 3.5 places the switch in the on state, whilst pH 8 switches it off;

Figure 5(b) shows the switching behaviour expressed as a series of consecutive peak resonances retraced at intervals of 15 seconds;

Figure 6(a) shows a pH calibration curve tracing the incremental resonance envelope changes between the on-state at pH 3.5 and the off-state at pH 8;

Figure 6(b) shows the variation in the amplitude and Q-factor of the resonance as a function of solution pH

Figure 7 shows the reversible binding that occurs between a boronic acid and cw-diols in aqueous media;

Figure 8 shows the chemical structure of a poly(acrylamide-co-3APB) copolymer; Figure 9 shows the pH curve of poly(acrylamide-co-3-APB);

Figure 10 shows the amplitude of response of MARS coated with a poly(acrylamide-co-3- APB) film as a function of glucose concentration;

Figure 1 1 (a) shows the response of MARS with poly(acrylamide-co-3-APB) film to the glucose solutions;

Figure 1 1 (b) shows the responses of the MARS as a function of glucose concentrations;

Figure 12 shows the variation of amplitude change of the MARS poly(acrylamide-co-3-APB) films to 15mM glucose; Figure 13 shows a plot of amplitude change of the MARS with poly(acrylamide-co-3-APB) films for a pH change of 4 to 10 against various 3-APB concentrations;

Figure 14 shows AQ between the response of the poly(acrylamide-co-3-APB) sensor in a pure pH 7.4 buffer and that in a l OmM glucose buffer solution at pH 7.4 as a function of the mol% 3-APB;

Figure 15 shows the amplitude difference between the signal from the MARS against concentration of cross-linker MBA when glucose incrementation is incrementally increased;

Figure 16(a) shows the amplitude of the response of the MARS of the poly(acrylamide-co-3- APB) film as a function of glucose concentration;

Figure 16(b) shows that amplitude of the response is proportional to the glucose concentration when it is not more than 7.5mM;

Figure 17 shows the response time of the MARS with a poly(acrylamide-co-3-APB) film to concentration change of glucose continuously from 0 to lOOmM at pH7.4;

Figure 18 shows the responses of the 3-APB based MARS to 5mM fructose, glucose, galactose and mannose solutions; and

Figure 19 shows plots of changes in amplitude of the MARS sensor as a function at mol% 3- APB with (a) apparent pK_a, (b) hydrophobscity (measured as ΔΑ₄) and (c) visco elasticity (measured as AQ).

DETAILED DESCRIPTION

Example 1

A novel chemically activated mechanical switch based on a composite acoustic resonator is described herein. The device is illustrated in Figure 1. A planar spiral coil is provided to excite electromagnetically the harmonics of a composite resonator made from a quartz disc and an acoustically thick hydrogel film coating based on a pH-sensitive hydroxyethyl methacrylate- co-methacrylate copolymer. A coaxial connection is provided between the spiral coil and signal generator, AM detector and lock-in amplifier. The detector employs a differential diode demodulation circuit to subtract the radio frequency signal returned from the coil from the larger excitation signal produced by the signal generator. To recover the frequency, amplitude and Q factor of the composite resonant device, LabView™ software, was used to process the raw amplitude and frequency data points collected from the spectrum.

Measurement of the acoustic resonance characteristics up to 100MHz was conducted with the above-described coil-based excitation system, which is capable of maintaining strong signal amplitudes across multiple harmonic frequencies. The operating frequency is selected from the signal generator front panel and used to recover differences in the acoustic envelope of each harmonic resonance, which in turn relates to the chemical properties of the hydrogel.

Switching action was observed in a 2.1 μιη thick hydrogel film deposited on 250 μιη thick AT quartz discs, when excited at the fundamental frequency (6.6 MHz) at pH 3.5, and quenched at pH 8 with an on/off amplitude ratio of 500:1. The switching point was found to be well defined by the argument of the hyperbolic tangent function of the wavevector (k)-film thickness (t) product (Tanh[kt]), and thus it was possible to measure the film elasticity as 5.8 x 10⁵ N/m² and the shear wave velocity as 24 ms^"1, without the use of multi -parameter fitting regimes. The approach allows both substantial amplification of the swelling response of the hydrogel and determination of the mechanical properties of the films. Thus an acoustic shear wave switch based on pH sensitive hydrogels has been provided.

In order to maximise the signal response and substantially modify the resonance in the manner described, it is desirable to have control of the resonance frequency of the probe. Similarly, for ease of handling and adherence of a hydrogel film, it is desirable to have a glass instead of a metal electrode surface that can be conveniently treated chemically. Thus, for technical reasons, the above-described tuneable device based on a simple coil and bare quartz substrate is advantageous. Many hydrogel materials and thicknesses can be utilized with this versatile format.

In the embodiment described herein, the mechanical behaviour of ionisable poly-HEMA hydrogels as a function of pH is utilized, when placed adjacent to the multi frequency quartz crystal resonator. The electrical features of the multi-frequency resonator and its detection characteristics may be evaluated. Understanding the formation of the hydrogel film and its interaction with the acoustic wave is another important aspect of the behaviour of the system. Accordingly, in the approach used here, phase contrast images of the deposited hydrogels were obtained and their thicknesses estimated. In addition, the behaviour of the system is explained in terms of the selected operating frequency and the acoustic impedance of the hydrogel resonator device has been estimated as a function of the pH value.

It is difficult to determine the thickness, density, elasticity and viscosity of the hydrogel films, since these parameters are largely unknown. For example, thickness determination is difficult owing to the soft nature of the hydrogel and its potential variation in thickness across the disc (however, it is possible to measure the thickness of the film as described later). Instead, it may be more pertinent to identify whether elasticity is a dominant factor for the likely range of thicknesses, and more specifically, whether the physico-chemical conditions that lead to the acoustic wave being switched on and off can be established. It should be noted that the system is more complex in some respects than those incorporating thin molecular films, as the thickness is a significant fraction of the acoustic wavelength. Thickness changes produce acoustic phase differences across the film that substantially affects the overall behaviour of the system. The wideband acoustic measurements of the hydrogel-disc composite made according to the present embodiment employ a readily available signal generator and lock-in amplifier, which are the same components previously employed for viscous fluid and protein adsorption studies. The spiral coil is hand wound, whilst the diode detector is constructed in-house.

Once the acoustic field is induced in the crystal by electrostriction, the acoustic behaviour is connected to the hydration state of the polymer, which in turn depends on the pH of the bathing medium. Acoustic coupling of the crystal to the thicker hydrogel films that are elastically "stiff has been measured, and the chemically relevant elasticity has been related to the switching behaviour itself.

EXAMPLES

2-Hydroxyethyl methacrylate (HEMA; 97%), ethylene dimethacrylate (EDMA - also known as ethylene glycol dimethacrylate), methacrylic acid (MAA), dimethoxyphenylacetophenone (DMAP), propan-2-ol, and (methacryloxypropyl)triethoxysilane were supplied by Aldrich Chemical Co. (Gillingham, U.K.). All other chemicals were of analytical grade and were supplied by Sigma or Aldrich. Aluminized 100 μιη-thick polyester film (grade MET401) was purchased from HiFi Industrial FilmLtd. (Stevenage, U.K.).

Composite resonator preparation

The Monomer mix was prepared from HEMA (89 mol.%), EDMA (cross-linker; 5 mol.%), MAA (6 mol.%) and an equal volume of propan-2-ol. The photoinitiator dimethoxyphenylacetophenone (DMAP) was then added to a final concentration of 1% (w/v). Crystals were treated overnight with 1% (w/v) 3 trimethoxysilylpropylmethacrylate in acetone. An aluminised polyester sheet was laid onto a flat glass plate and approximately 5 μΐ of monomer mix was pipetted to create a fluid 'bead'. The treated quartz disc was laid on top of the monomer mix and pressed to coat its underside evenly with polymer. UV- induced free radical polymerisation was carried out by 15 min exposure to UV radiation from a UV exposure unit (~350 nm). After polymerisation, the disc was peeled from the aluminised side of the polyester sheet, which acts as a release layer for the hydrogel-quartz composite. The final step was to rinse the composite thoroughly in methanol.

Spiral coils

Planar spiral coils with a DC resistance of 1 ohm, inductance of 0.5 mH and overall diameter of 5 mm were prepared from a 0.085 mm enameled copper wire obtained from RS Electronics (UK) and bonded to a 0.25 mm thick epoxy laminate board with a thin layer of cyanoacrylate adhesive. In order to design effective hydrogel sensors, it is necessary to understand how changes in their acoustic impedance derived from the detected electrical signals relate to the chemical- induced changes in the motion of the hydrogel.

Hydrogel-disc acoustic impedance

The system under consideration is a composite resonator consisting of a resonant plate attached to an additional hydrogel resonator of unknown thickness, density and viscoelasticity. In addition, there is a third liquid layer that is exposed to the upper surface of the composite. The acoustic impedance of the composite system can be conveniently described by (1-4):

G_w = 2nf co (3)

where k and t are the respective wavevector and thicknesses of the layers, and Z_q, Z_f and G_w are the quartz, hydrogel and water impedances. The combination of these impedances, Z_c, characterises the composite resonator and differs from the resonator alone by the 2^nd term of equation 2. This analysis shows that the hydrogel impedance Z_f can change dramatically when the argument kt of the hyperbolic tangent function is π/2, i.e. when chemical perturbation in the hydrogel matrix gives rise to very large shifts in acoustic amplitude and frequency. As the hydrogel dimensions are not as well defined as the quartz plate, and display much greater dimensional and mechanical variability, it is anticipated that film resonance only occurs in the lower harmonics, where the acoustic wave front remains planar and parallel. Hydro gel chemistry

The acoustic impedance of the hydrogel arises from the aggregate internal forces generated by a hydrophilic cross-linked copolymer layer that absorbs water from solution. The presence of weakly acidic carboxyl groups causes the gel to undergo changes in hydration as a function of the pH of the bathing medium [Montheard, J.P., M. Chatzopoulos, and D. Chappard, 2-Eydroxy thyl Methacrylate (Hema) - Chemical-Properties and Applications in Biomedical Fields. Journal of Macromolecular Science-Reviews in Macromolecular Chemistry and Physics, 1992. C32(l): p. 1-34]. Interestingly, the thickness and elasticity change simultaneously and continuously with pH. Under these circumstances, pH changes the thickness t because of hydration, and k because the elasticity is known to change with pH according to k=oW(pc) [Johnson, B., et al. Mechanical properties of a pH sensitive hydrogel. in SEM Annual conference proceedings. 2002. Milwaukee, WI]. This modifies the argument kt of the acoustic impedance of the film (1) and significantly amplifies the pH change acoustically through the hyperbolic tangent.

Detection of hydrogel disc composite

The special feature of the acoustic configuration employed in this work [Stevenson, A.C. and C.R. Lowe, Noncontact excitation of high Q acoustic resonances in glass plates. Applied Physics Letters, 1998. 73(4): p. 447-449] [Sindi, H.S., A.C. Stevenson, and C.R. Lowe, A strategy for chemical sensing based on frequency tunable acoustic devices. Analytical Chemistry, 2001. 73(7): p. 1577-1586] is the coupling to the disc across an air gap, such that there are no mechanical or electrical disturbances applied during the excitation/detection process. It is field driven and detected with the spiral coil, unlike conventional hard-wired acoustic devices. The process can be regarded as comprising of two simultaneous activities. The generation process whereby radio-frequency currents sourced from the signal generator circulate in the turnings of the coil. These produce an adjacent radio-frequency magnetic field, which itself supports a radio-frequency electric field. This field distribution which is dependent on the coil geometry, acts on dielectric materials at their dielectric boundaries to induce surface current [Stevenson, A.C. and C.R. Lowe, Magnetic-acoustic-resonator sensors (MARS): a new sensing methodology. Sensors and Actuators a-Physical, 1999. 72(1): p. 32-37] or charge [Stevenson, A.C, et al., The acoustic spectrophonometer: A novel bioanalytical technique based on multifrequency acoustic devices. Analyst, 2003. 128(10): p. 1222-1227]. Unequal boundaries, such as the top and lower surfaces of the piezoelectric crystal, which are exposed to water and air respectively, localise differential charge, much like the charge applied by electrodes to drive the crystal to motion [Stevenson, A.C., et al., Hypersonic evanescent waves generated with a planar spiral coil. Analyst, 2003. 128(9): p. 1175-1180], [Thompson, M., et al., Electromagnetic excitation of high frequency acoustic waves and detection in the liquid phase. Analyst, 2003. 128(8): p. 1048-1055]. The inverse process also applies, such that this motion in turn, leads to current induced back into the coil. The net result is that changes in the chemistry of the hydrogel are immediately transduced as mechanical changes, which affect the acoustic shear vibration, and, for the reasons mentioned above, the measured electrical signal.

Electrical impedance

An alternative way of viewing the system is to imagine that the amplitude of the electrical signal is dependent on the transfer function that relates the electromagnetic field distribution associated with the coil and the surrounding dielectric. The frequency shift and change in acoustic Q factor can be determined directly from equation (1) and related to the measured electrical Q factors and frequency shifts.

Assuming that thick film resonance conditions can be achieved, the model anticipates that substantial variation in the acoustic impedance of the film will be observed when the thickness, frequency and elasticity of the film gives rise to a π/2 shift in kt. The conditions and protocols whereby a chemically induced transition between zero resonance and thick film resonance is obtained are of particular interest.

Detection of disc-hydrogel resonance

Inspection of the fundamental resonance of the plate in water, followed by attachment of the hydrogel led to a reduction in the frequency and Q of the resonator (Figure 1). A pH of 3.5 was chosen for both measurements as the hydrogel is more compact and stable under these conditions [Marshall, A.J., et al., PH-sensitive holographic sensors. Analytical Chemistry, 2003. 75(17): p. 4423-4431]. The Q factor falls on attachment of the hydrogel through the creation of additional energy loss processes in the hydrogel. The hydrogel matrix appears to be a more "lossy" acoustic material than would be expected from one with a relatively open shift structure 'waggling' in a surrounding fluid. The resulting damping would downshift the resonance frequency and reduce the Q factor. Alternatively, it is conceivable that variations in the thickness of the film, which are likely to be considerably greater than in the quartz disc, prevent phase coherence, and thus degrade the resonant Q factor.

Determination of Thickness

A phase contrast microscope was used to measure film thickness independently, since it is especially important for determining the resonance behaviour of the composite. The noninvasive nature of microscopy avoids any disruption of the hydrogel film, as may be experienced with a contact profilometer, such as a Dektac. However, it is important that the film to be measured is transparent, in order to use phase contrast microscopy for profilometry. The procedure entails imaging a step from the bare surface of the quartz crystal to the upper surface of the hydrogel. Effectively, a contour map of the film edge is rendered, where each contour is separated by a fixed spacing period (Figure 3a). Knowing the dielectric constant of the poly-HEMA copolymer to be 1.5, and the wavelength of the light source to be 600nm, it was possible to estimate the film thickness from the contour map to be 2.1 μιτι for pipetted volumes of 10 μΐ.

Four separate devices with different volumes of monomer mix were pipetted onto the crystal prior to cross-linking with the ultraviolet light in order to evaluate whether the films were acoustically thick according to the definition of Martin [Anal. Chem. 2000, 72, 141-149]. Figure 3b shows that the frequency shift between the bare- and hydrogel-coated resonant disc is approximately linear but is not related to the volume applied. The measured shifts were found to be greater than the Sauerbrey-based frequency shifts calculated from the optical thickness, indicating that the acoustic phase shift across the films is significant. These measurements confirm that the applied hydrogel films are "thick" films.

Hydrogel-disc spectra

In order to induce different acoustic phase shifts without varying the chemistry or the device geometry, the frequency was adjusted to recover a series of harmonics over a range of frequencies at pH 3.5 (Figure 2). In general, a rapid reduction in resonance amplitude with frequency was found, with harmonics above 100 MHz being difficult to resolve. This behaviour is ascribed to material damping and variation in the film thickness.

The effect of changing the pH from 3.5 to 8 on the first five harmonics was investigated, by noting any changes in the acoustic profile, including its frequency, amplitude and Q factor. At the fundamental frequency (the first graph of Figure 4), a downward shift in frequency typical of a film of increased thickness was observed. This response is explicable in terms of the hydration of the hydrogel matrix as the polymer-bound carboxylic groups become ionised, drawing in counter-ions and water, such that the polymer phase swells and expands in thickness at the higher pH value.

At the 33 MHz harmonic, no frequency shift was apparent, since the resonance peak disappeared completely (the second graph of Figure 4). However, at higher frequencies, the shift pattern was reversed (the third graph of Figure 4). The frequency associated with the thicker hydrogel film subject to pH 8 was up-shifted. This behaviour is believed to be an example of thick film behaviour when the acoustic phase shift across the film is sufficient to support resonance in the film itself. This characteristic can be explained by the film decoupling from the disc resonance, such that the dominant reflection boundaries supporting the standing wave resonance is between the lower face of the disc and the disc-hydrogel interface, rather than the disc and the hydro gel- water interface.

The above-described example illustrates that harmonics (and in particular the 33 MHz harmonic) can be switched. However, any frequency, the fundamental or harmonics, for a suitable acoustically thick sensor layer could be used to realise switching behaviour.

Switching action

For a composite resonator with a thick film, the switching action is the most interesting feature. In this case, the transition between the on and off states is of a very high contrast. An example of good switching behaviour of a fundamental frequency is illustrated in Figure 5a in which the peak disappears completely on changing the pH from 3.5 to 8. Figure 5b illustrates that switching is reversible. The peak of the resonance in this example was repeatedly collected in a successive time-wise manner. The left-hand side of the trace represents the repeated collection of the same peak with time. Addition of pH 8 buffer, quenches the peak completely in about 10 seconds, whilst re-exposure to pH 3.5, recovers the original peak, proving that the elastic behaviour of the hydrogel-disc composite is reversible. The reason the switching action can be obtained in the first place is attributable to the very short acoustic wavelengths, which arise in the film due to a low shear wave velocity, estimated to be approximately 24 ms^"1. This value compares with values of 15 ms^"1 measured independently for poly-HEMA hydrogels. In addition, since the acoustic wavelength in the hydrogel at the switching point is well defined, it is possible to calculate the elasticity of the poly-HEMA at pH 8 from c=p.v² to be 5.8 x l 0⁵ N/m², which is consistent with the mechanical properties of 'soft' polymeric materials.

Calibration curve

A composite resonator was subject to incremental changes in pH in order to define the various stages of the switching transition. It switched reversibly, with complete cessation of the resonance at pH 8. These changes in profile of the acoustic resonance, and the variation in amplitude with pH, are shown in Figure 6a. The corresponding changes in the resonance amplitude and Q-factor are shown in Figure 6b. The data show substantial damping of the resonance, suggesting energy is being withdrawn from the quartz element by internal absorption in the hydrogel. The position of the turning point in the Q-factor and amplitude acoustic profiles can be anticipated by substituting different ionisable groups, as has been demonstrated with diffraction gratings reported previously.

A chemically actuated mechanical switch can be constructed from a composite resonator comprising a quartz disc and an overlaid hydrogel film. This switching action occurs when the film operates under thick film resonance conditions, where the acoustic phase is 90 degrees, in a hydrogel film 2.1 μηι thick, when the appropriate frequency with a multiple frequency acoustic device was employed. Sensitivity to elastic changes in the hydrogel was also significantly enhanced under switching conditions. Furthermore, not only can the technique assist the choice of appropriate monomer, cross-linker and copolymer ratios to give mechanically responsive acoustic or optical hydrogels, but if the film is thinner and probed at higher frequencies, it can also function as a chemically switchable component for investigating the dynamics of proteinacious films entrained in acoustic fields. Example 2

A glucose sensor capable of remote monitoring of glucose concentration is described herein.

A glucose-sensitive copolymer of 3 -acrylamido phenyl boronic acid (3-APB) and acrylamide with Ν,Ν-methylenebis acrylamido (MBA) as a cross-linker was synthesized through UV induced radical polymerization at room temperature. The pK_a of a poly(acrylamide-co-3- APB) is around 8.6, but the glucose binding reduces the pK_a value of the complex and the copolymer can measure glucose concentrations at physiological pH.

The main approach described here is to integrate a glucose-responsive poly(acrylamide-co-3- APB) film with into a magnetic acoustic resonator sensor (MARS) and investigate the response of the sensing apparatus to the variation of the glucose concentration in solution.

Complexation of Boronic acid with cis-diols

The reversible binding between boronic acids and cw-diols is shown in Figure 7 and is pH dependent. The chemical structure of the poly(acrylamide-co-3-APB) is shown in Figure 8. At low pH, the boronic acid is in a trigonal uncharged state and converts to a tetrahedral state when it is charged and the pH is increased. In the tetrahedral form, the boronic acid has a higher affinity for czs-diols than in the trigonal form, whose complex with cz^'s-diols is more easily hydrolyzed. Glucose contains a cw-diol structure. The two potential mechanisms by which glucose binds to a polymer-bound phenyl boronic group include: 1 :1 monomer binding and 1 :2 cross-linking binding. Equimolar 1 : 1 binding causes swelling of the polymer film, whilst 1 :2 binding leads to shrinking.

Response to pH of the Poly(acrylamide-co-3-APB) Film

The behaviour of the polymer films at various pH values was investigated as poly(acrylamide-co-3-APB) is pH sensitive and the complex of glucose and the tetrahedral 3- APB is more stable than the trigonal one. The pH titration curve of the polymer is defined and measured and the pK_a value was also calculated from the pH curve. The process for the synthesis of the copolymer is almost the same as with poly(HEMA-co-MAA); the only difference is a 30 min exposure to the UV. The pre-polymer solution was composed of 5mol/l monomer in 500μ1 solvent (DMSO containing 2% (w/v) of the photo-initiator DMPA). The buffer system was acetic acid (pH 4, 4.5, 5, 5.25 and 5.5), MES (pH 5.75, 6, 6.5), phosphate (pH 7, 7.5, 8), Tris (pH 8.5), Bicine (pH 9), CHES (pH 10) and phosphate (pH 12). The concentration of the buffer was l OmM, and the ionic strength was fixed at 154mM with sodium chloride.

Poly(acrylamide-co-3-APB) is gradually charged with increasing pH value of the buffer. There is a Donnan potential between the polymer and the aqueous medium around it which drives water to enter into and swell the hydrogel. At the same time, an osmotic effect also results in the hydration of the polymer. Figure 9 shows the pH curve collected at 6MHz with a poly(acrylamide-co-3-APB) film comprising 78.5mol% acrylamide, 1.5mol% MBA and 20mol% 3-APB. The calculated apparent pK_a value is 8.47.

Response to Glucose of the MARS with Poly (aery lamide-co-3-APB)

Reduction in the Apparent pK_a Caused by Glucose Binding

Figure 9 shows the pH curve of poly(acrylamide-co-3-APB) containing 78.5mol% acrylamide, 1.5mol% MBA and 20mol% MAA at 6MHz; the apparent pK_a value calculated from the curve is 8.47. In Figure 9, the poly(acrylamide-co-3-APB) polymer is almost neutral at physiological pH 7.4, in which case -91.5% of the 3-APB pendants are in the trigonal configuration and bind weakly to glucose. A solution of 5mM D-(+)-glucose was prepared in the buffers listed above in order to access the response of the coated resonator to glucose solution at various pH values. The pH curve at 6MHz was negatively shifted, and the apparent pK_a value is reduced from 8.47 (Figure 9) to 8.02. In this case, at pH 7.4, a proportion, 24%, of the boronic acid groups have become anionic and tetrahedral.

As shown in Figure 7, the dissociation of the neutral 3-APB is determined by the concentration of the protons in the solution and the binding between the glucose and the tetrahedral 3-APB, since glucose is weakly bound to the trigonal 3-APB. Calibration of the Glucose Response of the MARS

The response of the glucose-responsive poly(acrylamide-co-3-APB) MARS to the concentration of glucose in a buffer solution has been calibrated with the same polymer shown in Figure 9. The glucose solutions were made up in lOmM PBS buffer at pH 7.4. The concentrations of glucose were in the range of 0-15mM in 2.5mM steps. Figure 10 shows the amplitude of the response of the MARS coated with a poly(acrylamide-co-3-APB) film (78.5mol% acrylamide, 1.5mol% MBA and 20mol% 3-APB) as a function of the glucose concentration; data of responses got from triple testing glucose solutions in the range of 0- 15mM at 20MHz. Figure 10 shows that the amplitude of the signal from the MARS is almost linearly proportional to the glucose concentration over the 0-15mM range. The error is more obvious when the glucose concentration rises above 7.5mM. The sensitivity of the glucose meter is ~87.07mV/mM. The larger error at higher glucose concentrations is conceivably due to the conversion between the 1 :1 binding mode and the 2:1 cross-linking between the boronic acid and the glucose molecules.

The data shown in Figure 10 was collected: The polymer coated on the quartz disc was washed three times with pure PBS buffer and re-run under the same conditions.

Figure 11(a) shows three runs of the response of the MARS with a poly(acrylamide-co-3- APB) film (78.5mol% acrylamide, 1.5mol% MBA and 20mol% 3-APB) to the glucose solutions (OmM to 15mM) from the PBS buffer at pH 7.4 at 20MHz; (b) The responses as a function of the glucose concentrations of three runs are drawn separately. In Figure 11(b) it is noted that the responses of the poly(acrylamide-co-3-APB) MARS to glucose solutions containing more than 1 OmM glucose are slightly larger at the following two run than the first run while the responses obtained from the second and the third run testing are similar. This might be due to the pore size of the polymer network being expanded by the glucose molecules at the first run and that the glucose gained access to the polymer easier and swelled the hydrogel more afterwards.

Optimization of the 3-APB Concentration in the Glucose Sensor

The MARS is sensitive enough to detect the subtle change in the glucose concentration in buffer solutions. The relation between the response (the amplitude of the signal) and the glucose concentration is approximately linear (Figure 10); however, in order to enhance the response, the composition of the polymer film was optimized.

Response to Glucose with Various 3-APB Concentrations

The responses of the polymers to glucose with various mol% 3-APB were investigated. The glucose solution system was the same as given above, with the amplitude change (ΔΑ) being defined as:

ΔΑ₀ ¹⁵ = Amplitude (15mM glucose) - Amplitude (OmM glucose) (20MHz)

Figure 12 shows the ΔΑ₀ ¹⁵ of the MARS with poly(acrylamide-co-3-APB) films to 15mM glucose against various 3-APB concentrations at 20MHz; three sample for each mol%. The curve of amplitude change (ΔΑ₀ ¹⁵) against the mol% 3APB appears to be bell shaped (Figure 12). The maximum response is observed at a ~20mol% of 3-APB, and is reduced as more 3- APB is incorporated into the polymer. The results in Figure 12 are related to the contribution of the ionization of the poly(acrylamide-co-3-APB) as a function of the mol% 3-APB and the accessibility of the glucose into the polymer network which is related to the pore size and the hydrophilic/hydrophobic properties of the polymer network. The ionization of the poly(acrylamide-co-3-APB) in the presence of various glucose concentrations depends on the apparent pK_a of the polymer films or the apparent pK_a shift due to the glucose. In Table 1, the apparent pK_a of the poly(acrylamide-co-3-APB) as a function of mol% 3-APB has been calculated (20MHz); the apparent pK_a does not vary significantly when the concentration of the 3-APB is <20mol% but increases at higher mol%. The higher apparent pK_a, i.e. lower degree of ionization of the poly (acrylamide-co-3 -APB) containing 3-APB at >20mol% may explain the decline in the ΔΑ₀ ¹⁵ response in Figure 12.

Table 1: The apparent pK_a values of the poly(acrylamide-co-3-APB) with various 3- APB concentrations (from three samples) at 46MHz.

Mol% 3-APB 5 10 15 20 25 30 pK_a±STD* 8.62±0.06 8.55±0.03 8.51±0.10 8.64±0.21 8.99±0.04 8.91+0.05

*STD: standard error. The pK_a value Shifts due to Glucose Binding

The 3 -APB concentration dependence of the apparent pK_a shift of the poly(acrylamide-co-3- APB) due to glucose binding was investigated, since the poly(acrylamide-co-3-APB) with a lower apparent pK_a contains more tetrahedral boronic acid and could capture more glucose at physiological pH values. Polymers with 15, 20, 25 and 30mol% 3-APB respectively were fabricated and pH-curves responding to pure buffers in the range pH 4-10 were collected. The pH curves were re-recorded in the presence of 5mM glucose. For the four polymer films, the pK_a shifts are -0.52, -0.32, -0.66 and -0.46 respectively.

Effect of mol% 3-APB on the Hydrophilicity/Hydrophobicity of the Poly(acrylamide-c0-3-APB)

The hydrophilicity/hydrophobicity of the 3-APB polymers could be analyzed by observing the amplitude change as a result of imbibing water as the pH is changed from 10 to 4. The amplitude (ΔΑ₄ ¹⁰) is defined as:

ΔΑ₄ ¹⁰ = Amplitude (pH 10) - Amplitude (pH 4) (20MHz)

Figure 13 shows the ΔΑ₄ ¹⁰ of the MARS with poly(acrylamide-co-3-APB) films to pH change (pH 4-10) against various 3-APB concentrations at 20MHz; three samples were assessed for each mol%. pH responses of the polymers with various 3-APB concentrations (5-30mol% with 5mol% steps) were collected. Figure 13 illustrates the amplitude change as a function of the percentage of 3-APB in the copolymer. The amplitude change is maximal when the polymer contains 20mol%; when the mol% 3-APB is >20mol%, the polymer film swells less. Theoretically, increasing the density of 3-APB raises the Donnan potential and the osmotic effect and more water will enter the polymer network. However, the hydrophobicity of 3-APB hinders the water and glucose simultaneously from entering the hydrogel. Hence, at 20mol%, the effects of osmosis and hydrophobicity arrive at a balance. Therefore, it is proposed that the hydrophilic/hydrophobic property of the poly(acrylamide- co-3-APB) is a predominant factor determining the sensitivity of the acoustic glucose sensor. Effect of mol% 3-APB on the Viscoelasticity of the Polymer

The swelling ability of the poly(acrylamide-co-3-APB) copolymer also affects the penetration of the glucose molecules into the polymer and is qualitatively defined by the Q-factor of the polymer-coated sensor. As discussed above, the poly(acrylamide-co-3-APB) copolymer behaves as an elastic material at the high operating frequencies of the MARS; hydrogen bonding with water or glucose molecules might enhance the elasticity of the polymer and increase the Q-factor.

The Q-factors of the poly(acrylamide-co-3-APB) sensors increase proportionally with the operating frequency <~140MHz and subsequently saturate above 140MHz; for example, the Q-factor at the saturation level of the polymer with 15mol% 3-APB is -3250 whilst that of the polymer with 20mol% is -2000 and the one with 30mol% is less swollen with a Q-factor of -900. The increase in the Q-factor as a function of the operating frequency is likely to be due to less acoustic energy penetrating and being lost in the polymer while the augmentation of the elasticity of the polymer might also contribute. At high frequencies, the acoustic energy is focused on the polymer-quartz interface and the viscoelastic property of the polymer reaches equilibrium, and therefore, the Q-factor is saturated at frequencies > 140MHz.

The swelling behaviours of the same polymers in lOmM glucose phosphate buffer in pH 7.4 were also observed. In this case, the saturation level of the corresponding polymer Q-factor containing 15mol% 3-APB is -4000, the polymer with 20mol% is -3000 and the one with 30mol% is less swollen with a Q-factor of -1500. The Q-factors' differences (AQ) of the polymers in a pure buffer and in a lOmM glucose solution are 750 for 15mol% 3-APB, 1000 for 20mol% 3-APB and 600 for 30mol% 3-APB. The AQ at the saturation levels as a function of the mol% 3-APB are shown in Figure 14, which shows AQ between the response of the poly(acrylamide-co-3-APB) sensor in a pure pH 7.4 buffer and that in a lOmM glucose buffer solution at pH 7.4 as a function of the mol% 3-APB. The results in Figure 14 indicate that glucose-binding enhances the elasticity of the polymer with an optimum observed at between 20 and 25mol% 3-APB. In addition to the hydrophilicity/hydrophobicity of the polymer, the swelling ability and the pore size of the polymer network also affect the sensitivity of the glucose sensor. The first half of the curve in Figure 12 reflects the fact that the sensitivity of the glucose polymer is proportional to the concentration of the phenyl boronic group and the second half implies that the sensitivity of the polymer is, in addition, determined by the hydrophilic/hydrophobic balance, the viscoelastic properties and the pore size of the network.

Optimization of mol% Cross-linking in the 3-APB Polymer

It is believed that the pore size and the viscoelasticity of the hydrogel network also affect the performance of the glucose sensor where steric hindrance may play a significant role.

According to the results above, the maximum response (ΔΑ₀ ¹⁵) was achieved when the concentration of 3-APB was 20 mol%. The effect of cross-linking was investigated by keeping the percentage of 3-APB as 20mol% and monitoring the responses of the polymers with 0.5, 1.5 and 2.5mol% MBA respectively (Figure 15). Figure 15 shows the amplitude difference between the signal from the MARS when the reference (glucose concentration is OmM) and glucose concentration is 15mM against the concentration of cross-linker, MBA. Figure 15 shows glucose buffer solutions were injected continuously in the concentration range of 0-15mM in steps of 2.5mM. The results in Figure 15 suggest that reducing the mol% of the cross-linker cannot help to enhance the sensitivity to glucose of the poly(acrylamide- CO-3-APB) sensor. AQ, is also employed to investigate the swelling behaviour of the poly(acrylamide-co-3-APB) copolymers with cross-linking in the range of 0.5-2.5mol% exposed to a 1 OmM glucose solution.

5.7 Detection of Hyperglycaemia and Hypoglycaemia

For diabetes patients, the blood glucose concentration rise to levels well above normal (~4.4mM), hyperglycaemia, while when sufferers have taken insulin improperly, the blood glucose level can fall below the normal range, hypoglycaemia. Therefore, the effective working range of the glucose sensor should be measured. Glucose solutions were prepared in pH 7.4 PBS buffer at concentrations of 0, 2.5, 5, 7.5, 10, 12.5, 15, 20, 40, 60, 80 and lOOmM. The polymer film consisted of 83.5mol% acrylamide, 1.5mol% MBA and 15mol% 3-APB.

Figure 16 shows (a) the amplitude of the response of the MARS with a poly(acrylamide-co-3- APB) film (83.5mol% acrylamide, 1.5mol% MBA and 15mol% 3-APB) as a function of the glucose concentration (0-lOOmM), fitted in an exponential equation by SigmaPlot; (b) the amplitude of the response is proportional to the glucose concentration when it is <7.5mM.

As shown in Figure 16, the response of the MARS as a function of the glucose concentration becomes saturated above 20mM. Consequently, the effective working range of the glucose sensor based on the MARS is 0-20mM, and below 7.5mM the amplitude of the response is largely proportional to the glucose concentration (Figure 16). The data given in Figure 16(a) have been fitted to an exponential equation which suggests that the response of the MARS sensor as a function of the glucose concentration might follow an exponential rise model. Figure 17 shows the response times to glucose concentrations in the range 0-l OOmM and that above 20mM glucose, the response time is <10min. Figure 17 shows the response time of the MARS with a poly(acrylamide-co-3-APB) film (83.5mol% acrylamide, 1.5mol% MBA, 15mol% 3-APB) to the concentration change of glucose continuously from OmM to lOOmM at pH 7.4.

The saturation level observed in Figure 16 at glucose concentration >20mM and the shrinking of the response time when the 3-APB content is >20mol% (Figure 17), might relate to the swelling of the polymer network. The hydrogel swells more in solutions containing higher glucose concentrations, which, in turn facilitates the penetration of the glucose solutions. However, since the poly(acrylamide-co-3-APB) copolymer is cross-linked, the swelling ability of the network is limited and, therefore, it is possible that the saturation level is being approached in Figure 16 corresponds to the swelling limitation and, in addition, the steric effect of the complex of glucose and the aromatic boronic acid. Effect of the Electric Field on Glucose Response

According to the study described above, the operating frequency affected the apparent pK_a value measured by the glucose sensor. The theory of polymer brushes and the notion of local pK_a were engaged in order to understand the performance of the glucose sensors over a wide acoustic spectrum. In this study, how the performance of the glucose sensor is influenced by the operating frequency is investigated by determining the effective detection range of the sensor.

The composition of the polymer was acrylamide 78.5mol%, cross-linker MBA 1.5mol%, 3- APB 20mol% and the responses of the MARS over the frequency range 6-73MHz as a function of glucose concentration are found to be quite similar to each other. Hence, the influence of the operating frequency on the response of the MARS with a poly(acrylamide- co-3-APB) film can be ignored.

Sugar Assay

The previous studies based on holographic sensors suggested that the glucose molecule binds to the phenyl boronic group in two ways: 1 : 1 monomer binding and 1 :2 cross-linking binding. Stoichiometric 1 :1 binding causes swelling of the polymer film, while 1 :2 stoichiometry leads to shrinking. There are five chemical structures of the glucose: One open- chain form and four cyclic forms which can in principal combine with the phenyl boronic acid.

Isomers of glucose were studied to understand the mechanism of the combination between the phenylboronic acid and the m-diols. Furthermore, apart from glucose, the most abundant sugars contained in blood are fructose and galactose. Fructose, galactose and mannose solutions (5mM) were made up in l OmM PBS buffer with 154mM ionic strength at physiological pH. The quartz disc with a poly(acrylamide-co-3-APB) copolymer film was equilibrated in the three sugar solutions, and in between measurement, the film was washed with phosphate buffer until the response returned to the reference level (phosphate buffer).

Figure 18 shows the chemical structures of β-D-fructofuranose, a-D-galactopyranose and β- D-mannopyrano se . The response of the polymer to fructose is substantially larger than to the other sugars, with an amplitude change compared with the reference of 362.95mV (Figure 18). The responses to glucose, galactose and mannose are smaller and relatively similar to each other. The ΔΑ of glucose is 52.25mV, that of galactose is 30.15mV and mannose is 15.2mV. The strong response to fructose can be explained by the affinity of the cw-diols to boronic acid and the proportions of the effective isomers. Fructose has been reported to have higher affinity for the boronic acid than the other three sugars.

Discussion and Conclusions

A synthetic phenylboronic acid, 3-acrylamidophenylboronic acid (3-APB), has been integrated into the MARS to create a selective sensor for glucose, which, because the MARS is stimulated wirelessly by a RF electric field constitutes a promising alternative for developing an implantable continuous glucose monitor. 3-APB is a derivative of phenylboronic acid and its copolymer with acrylamide can respond to glucose at physiological pH. By replacing the traditional enzyme glucose oxidase with synthetic phenylboronate ligands, the performance of the glucose sensor could be more durable and reversible, since denaturation of the enzyme reduces the stability and sensitivity of the sensor. A synthetic glucose responsive hydrogel is also considered able to realize an in vivo glucose- insulin closed loop system. The sensitivity, repeatability, selectivity and response time are crucial parameters for a glucose monitoring system. The response to glucose of the MARS with a poly(acrylamide-«?-3-APB) copolymer film has been studied together with the mechanism of the glucose binding in the polymer film.

The sensitivity to glucose of the MARS coated by a poly(acrylamide-co-3-APB) adlayer is determined mainly by the concentration of the 3-APB, the hydrophobicity and viscoelasticity of the polymer (Figure 19). Figure 19(a) shows the ΔΑ₀ ¹⁵ of the MARS sensor and the apparent pK_a value of the poly(acrylamide-co-3-APB) as a function of the mol% 3-APB; (b) The ΔΑο¹⁵ of the MARS sensor and the ΔΑ₄ ¹² of the poly(acrylamide-co-3-APB) as a function of the mol% 3-APB; (c) The ΔΑ₀ ¹⁵ of the MARS sensor and the AQ of the poly(acrylamide-co-3-APB) as a function of the mol% 3-APB.

In conclusion, a preliminary study on a novel glucose sensor system based on the MARS has been implemented. The change in viscoelastic properties inside poly(acrylamide-co-3-APB) networks caused by binding glucose could be measured by the MARS and the corresponding glucose level could be calibrated. The glucose monitoring system has a good reversibility and repeatability. The sensitivity of the sensor is on average ~87.07mV/mM. The results indicate that MARS could be a universal platform for investigating the physical properties of the very thin polymer films and developing smart polymer-based biosensors.

While this invention has been particularly shown and described with reference to preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the invention as defined by the appended claims.

Claims

1. A sensing apparatus comprising: a resonator, a sensor and a detector, the sensor being mechanically coupled to the resonator, the sensor comprising a sensor material which changes between a first state and a second state when exposed to a change in the surrounding environment, wherein the sensor is driven by the resonator and the detector is responsive to the change in state of the sensor material, wherein the sensor material is in the form of an acoustically thick layer.

2. A sensing apparatus according to claim 1, wherein the detector comprises an electromagnetic field generator capable of being arranged to direct an electromagnetic field towards the sensor.

3. A sensing apparatus according to claim 2, wherein the electromagnetic field generator and the detector comprise a common structural element for generating an electromagnetic field and detecting an electromagnetic field.

4. A sensing apparatus according to claim 2 or claim 3, wherein the electromagnetic field generator is tuneable.

5. A sensing apparatus according to any one of claims 2 to 4, wherein the electromagnetic field generator is a coil.

6. A sensing apparatus according to any one of claims 2 to 5, further comprising a signal generator and a lock-in amplifier connected to the electromagnetic field generator and the detector.

7. A sensing apparatus according to claim 6, wherein the detector comprises a differential diode demodulation circuit for subtracting a detected signal from a signal produced by the signal generator.

8. A sensing apparatus according to any preceding claim, comprising an impedance measuring circuit for measuring a change in impedance in the detector.

9. A sensing apparatus according to any preceding claim, wherein the resonator comprises a magnet capable of being arranged to direct a magnetic field towards the sensor.

10. A sensing apparatus according to any preceding claim, wherein the resonator comprises an exciter mechanically coupled to the sensor.

11. A sensing apparatus according to claim 10, wherein the exciter comprises a piezoelectric material.

12. A sensing apparatus according to claim 11, wherein the piezoelectric material is quartz.

13. A sensing apparatus according to any one of claims 1 1 and 12, wherein the piezoelectric material is in the form of a layer.

14. A sensing apparatus according to claim 13, wherein the piezoelectric layer is 50 to 1000 μηι thick.

15. A sensing apparatus according to any one of claims 10 to 14, wherein the exciter comprises a magnetostrictive material.

16. A sensing apparatus according to any one of claims 10 to 15, wherein the exciter comprises a metallic material.

17. A sensing apparatus according to claim 16, wherein the metallic material is in the form of a layer.

18. A sensing apparatus according to claim 17, wherein the metallic layer is 50 to 1000 μιη thick.

19. A sensing apparatus according to any preceding claim, wherein the acoustically thick layer is 0.1 μηι to 1 mm thick.

20. A sensing apparatus according to claim 19, wherein the acoustically thick layer is 0.1 μηι to 100 μηι thick.

21. A sensing apparatus according to claim 20, wherein the acoustically thick layer is 0.1 μηι to 10 μηι thick.

22. A sensing apparatus according to claim 21, wherein the acoustically thick layer is 0.5 μηι to 5 μηι thick.

23. A sensing apparatus according to any preceding claim, wherein the acoustically thick layer is more than one molecule thick.

24. A sensing apparatus according to any preceding claim, wherein the sensor material is a hydrogel.

25. A sensing apparatus according to claim 24, wherein the hydrogel is responsive to change in chemical environment.

26. A sensing apparatus according to claim 25, wherein the change in chemical environment is a change in pH.

27. A sensing apparatus according to any one of claims 24 to 26, wherein the hydrogel is hydroxyethyl methacrylate-co-methacrylate.

28. A sensing apparatus according to claim 25, wherein the change in chemical environment is a change in concentration of an analyte.

29. A sensing apparatus according to claim 28, wherein the analyte is a physiological analyte.

30. A sensing apparatus according to claim 28 or claim 29, wherein the analyte is a sugar.

31. A sensing apparatus according to claim 30, wherein the sugar is glucose.

32. A sensing apparatus according to any one of claims 28 to 31, wherein the hydrogel is a polymer containing phenyl boronic acid pendant groups.

33. A sensing apparatus according to claim 32, wherein the polymer is poly(acrylamide- co-3-acrylamido phenyl boronic acid) or poly(acrylamide-co-2-acrylamido phenyl boronic acid).

34. A sensing apparatus according to any preceding claim, wherein the sensing apparatus is a chemically actuated electro-mechanical sensing apparatus.

35. A sensing apparatus according to any one of claims 2 to 34, wherein the resonator and sensor are in the form of an implantable device for implanting into a human or animal subject; and wherein the detector is responsive to the sensor ex vivo.

36. Use of the sensing apparatus according to any preceding claim in a method of sensing.

37. Use of the sensing apparatus according to any one of claims 1 to 35 as a switch.

38. An implantable device for implanting into a human or animal subject, which comprises a resonator and a sensor in which the sensor is mechanically coupled to the resonator, as defined in any one of claims 2 to 35.

39. Use of a detector in a method of remote sensing of a physiological state of a subject implanted with the implantable device of claim 38, wherein the detector is responsive to the sensor material of the sensor.

40. Use according to claim 39, wherein the detector is as defined in any one of claims 2 to 7.

41. Use according to claim 39 or claim 40, wherein the sensing of the physiological state of the subject comprises sensing of an analyte.

42. Use according to claim 41, wherein the analyte is a sugar.

43. Use according to claim 42, wherein the sugar is glucose.

44. A switch comprising the sensing apparatus according to any one of claims 1 to 35.

45. A method of controlling a system based upon a change in the surrounding environment using a sensing apparatus according to any one of claims 1 to 35.

46. A method of making a composite probe comprising a resonator element and an acoustically thick hydrogel film thereon, the method comprising: preparing a monomer mixture for the hydrogel; applying the monomer mixture to a release layer; applying the resonator element to the monomer mixture; polymerising the monomer mixture to form the acoustically thick hydrogel film adhered to the resonator element thereby forming the composite probe; and peeling the composite probe from the release layer.

47. A method of making a composite probe according to claim46, wherein the resonator element is treated with a binding agent prior to application of the monomer mixture to promote binding of the hydrogel to the resonator element.

48. A method of making a composite probe according to claim 47, wherein the binding agent is trimethoxysilylpropylmethacrylate.

49. A method of making a composite probe according to any one of claims 46 to 48, wherein the release layer comprises an aluminised polyester sheet.

50. A method of making a composite probe according to any one of claims 46 to 49, wherein the resonator element is piezoelectric.

51. A method of making a composite probe according to any one of claims 46 to 50, wherein the hydrogel is prepared from a mixture of 2-hydroxyethyl methacrylate (HEMA), ethylene dimethyacrylate (EDMA), methacrylate (MAA) and a photoinitiator.

52. A method of making a composite probe according to claim 51, wherein the photoinitiator is dimethoxyphenylacetone.