Acoustic wave sensor based on the modal localisation principle and measurement method for use in the acoustic wave sensor based on the modal localisation principle

Abstract

The present invention falls within the industrial sector responsible for laboratory measurement equipment, and discloses an acoustic wave sensor, the measurement method for using the acoustic wave sensor based on the modal localisation principle, as well as the device implementing the method. The sensor improves the sensitivity, detection limit and robustness against external interferences. The invention allows multiple resonators to be integrated into a single device with only two contacts. Likewise, the invention enables the simultaneous detection of multiple different analytes in the same sample.

Description

[0001] DESCRIPTION

[0002] ACOUSTIC WAVE SENSOR BASED ON THE MODAL LOCALIZATION PRINCIPLE AND APPLICATION MEASUREMENT PROCEDURE IN THE ACOUSTIC WAVE SENSOR BASED ON THE MODAL LOCALIZATION PRINCIPLE

[0003] Technology sector

[0004] The present invention falls within the industrial sector responsible for laboratory equipment for measurement, especially in sectors such as medical diagnosis, food safety or environmental, using a shear resonator.

[0005] Object

[0006] Improve sensitivity, detection limit, and robustness to external interference. Develop lightweight, portable instruments suitable for use outside the laboratory. Eliminate the need for thermoregulatory elements such as Peltier cells. Integrate multiple resonators into a single device with a simple connection interface based on just two electrical contacts. Enable the simultaneous detection of multiple different analytes in the same sample. Eliminate the need for additional meters to monitor interfering external factors and improve measurement stability.

[0007] Background

[0008] Acoustic wave resonators based on the piezoelectric effect are devices used to accurately measure both variations in mass per unit area and the viscoelastic properties and conformational changes of the media deposited on them, through the changes that the complex resonance frequency of these crystals undergoes when operating as resonators.

[0009] Among acoustic wave resonators, the quartz crystal microbalance (QCM) stands out as a label-free, direct detection tool suitable for real-time monitoring. The QCM operates using a gravimetric technique that relates changes in mass on the sensor surface to shifts in the resonant frequency, Afr. This approach has been widely applied in biochemical detection, including immunoassays, protein adsorption, and DNA hybridization.

[0010] These devices are ideal for monitoring physicochemical processes in situ, but environmental conditions such as temperature, humidity, pressure, and flow rate can cause drift and variations in the sensor's response. For this reason, it is necessary to keep constant all factors that could affect the sensor's stability by masking the signals of interest. This is no easy task and often increases the complexity and cost of the testing equipment, frequently hindering the development of lightweight and portable instruments that can be used in applications outside the laboratory.

[0011] For example, in the case of temperature, active thermal control systems are used, usually based on the inclusion of thermoregulatory elements, such as Peltier cells, to control the temperature of the sample and the sensor during the experiment.

[0012] Numerous authors have tried to find alternative solutions to the use of expensive and complex control systems. One of the first approaches was based on the use of two resonators, one of which is used as a reference (VM Mecea, JO Carlsson, P. Heszler, and M. Bártan, “Development and testing of a high temperature quartz crystal microbalance,” Vacuum, vol. 46, no. 7, pp. 691–694, Jul. 1995, doi: 10.1016 / 0042-207X(94) 00147-2). By subtracting the frequency variations of both resonators, it is possible to cancel the effect of temperature fluctuations on the measurement. This method works well when two conditions are met: 1) the temperature is the same in both resonators, and 2) the temperature response of both resonators is the same. These two conditions are not always met. If this method is applied, proper thermal design of both the instrument and the resonators is necessary.

[0013] Rahtu et al. A. Rahtu and M. Rítala, “Compensation of temperature effects in quartz crystal microbalance measurements,” Applied Physics Letters, vol. 80, no. 3, pp. 521–523, 2002, doi: 10.1063 / 1.1433904, compare the performance of the previous method with an alternative approach based on modeling the frequency variation produced by temperature changes using a third-degree polynomial. Through a numerical fit of the experimental data to this model, thermal effects are corrected without the need for an additional reference sensor. The results obtained with this method are comparable to those obtained using the reference sensor (Rahtu et al.). This approach is only valid for monotonic temperature variations and may have convergence problems for experiments lasting longer than 15 minutes. Furthermore, a temperature sensor must be included in the instrument.

[0014] Other authors (DE Pierce, “A temperature insensitive quartz microbalance,” IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 45, no. 5, pp. 1238–1245, 1998, doi: 10.1109 / 58.726449) propose canceling the temperature effect by using SC-cut quartz acoustic resonators. These resonators are specifically designed to compensate for mechanical stress, and their temperature response is independent of the temperature rate. By simultaneously measuring two harmonic modes of the resonator and assuming a gravimetric operating regime, it is possible to calibrate the crystal's temperature response without the need for an external thermometer. The main disadvantage of SC-cut resonators is their higher cost due to the complex double-rotation process required for their manufacture.Furthermore, thermal compensation based on this method does not apply to dissipation measurements or to experiments where the thickness of the layer deposited on the sensor or its viscoelastic or hydrodynamic properties do not ensure a gravimetric working regime.

[0015] In addition to temperature, other factors such as pressures and mechanical stresses due to the measuring cell, aging of the resonator, external vibrations, non-specific surface adsorption, presence of impurities or dislocations in the quartz, or humidity can affect the stability of the resonance frequency and the quality factor, making the measurement difficult and degrading the detection limit of the technique.

[0016] In these cases, the state of the art refers to different methods that attempt to correct drift in a "blind" manner, regardless of the mechanism that produces it in individual sensors. Most of them are based on calibrating the sensor using a reference sample, which is then used to compensate for the baseline by applying an additive or multiplicative correction afterward. These methods cannot be applied in real time and introduce the need for a calibration phase in the measurement. Other methods are based on frequency analysis to filter out components due to electronic noise or drift from the signal of interest by applying averaging, smoothing, or filtering. They generally work well for reducing high-frequency fluctuations (noise), but they do not yield good results in the case of low-frequency signals (drift), whose spectrum often overlaps with the components of interest, making filtering difficult.

[0017] Other authors have proposed the use of statistical tools to improve frequency stability. One such method is Principal Component Analysis (PCA). It is based on the dimensional reduction of a large dataset to highlight its most statistically significant components, which ideally are related to the signals of interest, while eliminating the less significant components, which could be associated with frequency instability. Recently, PCA has been used to improve the detection limit (E. Corradi et al., “An objective, principal-component-analysis (PCA) based, method which improves the quartz-crystal-microbalance (QCM) sensing performance,” Sensors and Actuators, A: Physical, vol. 315, p. 112323, 2020, doi: 10.1016 / j.sna.2020.112323.) of acoustic sensors in measurements with multiple harmonics.Assuming a gravimetric operating regime, the measurements of all modes are reduced to a single measurement with an improved detection limit. The combined use of PCA and neural networks has also been reported to eliminate the humidity effect on toluene gas measurements (B. Mumyakmaz, A. Ózmen, M.A. Ebeoglu, C. Ta§altin, and I. Gürol, “A study on the development of a compensation method for humidity effect in QCM sensor responses,” Sensors and Actuators, B: Chemical, vol. 147, no. 1, pp. 277–282, 2010, doi: 10.1016 / j.snb.2010.03.019) performed by an array of QCM sensors. Due to their high computational cost, these PCA-based statistical methods are typically used for measurement classification rather than for real-time signal correction.

[0018] An alternative approach to traditional resonant frequency monitoring relies on the physical phenomenon known as modal localization. When two identical resonators are mechanically coupled, their oscillations synchronize. This results in two resonant modes very close to each other (eigenstates): one symmetric, in which the oscillators move in the same direction, and the other antisymmetric, in which they move in opposite directions. Spletzer et al., based on the concept of mode localization presented by PW Anderson (PW Anderson, “Absence of Diffusion in Certain Random Lattices,” Phys. Rev., vol. 109, no. 5, pp. 1492-1505, Mar. 1958, doi: 10.1103 / PhysRev. 109. 1492.), showed how these coupled resonators could be used to increase the mass sensitivity of microcantilevers (M. Spletzer, A. Raman, AQ \Nu, X. Xu, and R.Reifenberger, “Ultrasensitive mass sensing using mode localization in coupled microcantilevers,” Appl Phys Lett, vol. 88, no. 25, pp. 2004–2007, 2006, doi: 10.1063 / 1.2216889.) A small perturbation in one of the resonators, such as the added mass of a target analyte, breaks the symmetry and results in a shift of the two eigenstates that is several orders of magnitude greater than the shifts in the resonance frequency of the individual resonators that are traditionally monitored. The shifts in the eigenstates are greater the lower the coupling level. Another advantage of this approach is that it exhibits an inherent common-mode rejection ratio (CMRR), since the eigenstate shifts are insensitive to external noise affecting both resonators (P. Thiruvenkatanathan, J. Yan, and AA.Seshia, “Common mode rejection in electrically coupled MEMS resonators utilizing mode localization for sensor applications,” 2009 IEEE International Frequency Control Symposium Joint with the 22nd European Frequency and Time Forum, pp. 358-363, 2009, doi: 10.1109 / FREQ.2009.5168201.). Common mode rejection characterizes the ability of a device's electromechanical response to remain stable despite changes in ambient measurement conditions (temperature, pressure, humidity, etc.), thus demonstrating its robustness against external factors that could affect the measurement.

[0019] The operating principle of modal localization is relatively independent of the quality factor (Q) of the coupled resonators, which facilitates their manufacturing process (P. Thiruvenkatanathan, J. Woodhouse, J. Yan, and AA Seshia, “Limits to mode-localized sensing using micro- and nanomechanical resonator arrays,” J Appl Phys, vol. 109, no. 10, p. 104903, May 2011, doi: 10.1063 / 1.3590143.).

[0020] Recently, Humbert et al. applied the concept of modal localization to QCM sensors using electrical coupling artificially introduced into the system via a negative feedback loop implemented with an FPGA-based circuit (C. Humbert, V. Walter, and T. Leblois, “A mass sensor based on digitally coupled and balanced quartz resonators using mode localization,” Sens Actuators A Phys, vol. 335, no. xxxx, p. 113378, Mar. 2022, doi: 10.1016 / j.sna.2022.113378). Although their results show an improvement in mass sensitivity, they do not confirm the robustness of the method to temperature changes. Furthermore, they require an external electronic circuit, which increases the cost and complexity of the solution.

[0021] The applicant is unaware of any technical solutions that would solve the problem in a manner as advantageous as the proposed invention.

[0022] Description of the invention

[0023] The present invention proposes a novel acoustic wave sensor, which we call the CTSM (Coupled Thickness Shear Mode) sensor. This sensor, based on the modal localization phenomenon, improves the stability and mass limit of detection (LoD) of conventional QCM sensors based on the Sauerbrey resonance frequency measurement, resolving one or more of the drawbacks described above.

[0024] The CTSM sensor consists of two or more acoustic wave resonators embedded in the same AT-cut quartz substrate, with a thickness ranging from 10 to 500 micrometers. These resonators vibrate in shear mode. Each resonator has two metal electrodes, deposited between 5 and 200 nanometers thick, superimposed on the two faces of the quartz substrate. The resonators exhibit weak coupling due to the mechanical energy exchanged between them via shear acoustic waves (Thickness Shear, TS), with a coupling constant K less than 0.09. Achieving this acoustic coupling requires a specific design of the resonator geometry and electrode thickness, as their proximity and effective surface area determine the coupling level and, consequently, the sensor's sensitivity.For example, for resonant frequencies on the order of 5 MHz, the distance between electrodes should not exceed 3 mm and their surface area should be at least 25 mm. 2 Only one of the resonators, which we call 'active', is electrically excited, while the others, called 'passive', are allowed to vibrate freely due to acoustic coupling. Therefore, the interconnection system required for its operation has only two electrical contacts. The electrodes of the passive resonators can be left disconnected in an open circuit, or connected to each other by a short circuit.

[0025] The response of CTSM sensors exhibits an electrical impedance spectrum with as many minima around the resonant frequency as there are resonators fabricated on the quartz substrate. Each of these impedance minima corresponds to a coupled mode of the resonator system.

[0026] To characterize the CTSM sensor's response, instead of measuring variations in its resonant frequency(ies) as in the traditional approach, we propose measuring the amplitude of its coupled modes. To this end, we first monitor the impedance spectrum around the CTSM sensor's nominal resonant frequency and then identify its coupled modes as the local maxima in the conductance spectrum. For example, if two coupled resonators are integrated into the CTSM sensor, two local maxima (G1 and G2) will appear in its spectrum. According to our results, the variation in the modal amplitude difference (AG = GI - G2) and its ratio (R = GI / G2) depend only on the mass difference of the coatings deposited on the surface of the two resonators.Therefore, if we use the 'active' resonator as a reference, it is possible to determine the characteristics of the sample deposited on the 'passive' resonator with an Instrumental Detection Limit (IDL) several orders of magnitude greater than the measurement of changes in the resonance frequency of a conventional QCM sensor, thus solving the technical problem of the low sensitivity of traditional acoustic sensors.

[0027] Furthermore, as predicted by theory, the CTSM sensor's response exhibits a Common Mode Rejection Ratio (CMRR) several orders of magnitude better than that shown by the resonant frequency measurement of conventional quartz microbalance (QCM) sensors. Therefore, this response is highly robust against external factors (humidity, temperature, pressure, vibration, etc.) that could interfere with the measurement if these factors affect both resonators equally. This resolves the technical problem associated with the sensor's dependence on these external factors, a characteristic of conventional QCM sensors. The CTSM sensor does not require additional electronic circuitry for modal localization. Nor does it need external meters to monitor interfering factors to improve measurement stability.The CTSM sensor has a high level of immunity not only to one of these interfering factors but to the simultaneous effect of all of them.

[0028] The CTSM sensor does not require the use of a reference sample for baseline instabilities correction.

[0029] The CTSM sensor provides a tool for real-time monitoring of a chemical or physical process whose outcome can be evaluated in terms of the mass variation of a coating deposited on the surface of said sensor in a gaseous medium or in a vacuum.

[0030] When the CTSM sensor consists of more than two resonators, it can be used to simultaneously monitor different samples deposited on each of the 'passive' resonators. In this way, the CTSM can function as a sensor array but with a much simpler and more robust connection interface based on only two electrical contacts (the contacts of the active resonator). Due to this unique feature, the cells and measuring instruments commonly used with commercial QCM sensors are fully compatible with the proposed CTSM sensor design.

[0031] To use CTSM sensors to determine the surface mass density of a coating deposited on a passive resonator, the following procedure must be followed. First, continuously monitor the electrical impedance of the active resonator around its nominal resonant frequency with a sampling period of 0.01 to 1 second and a bandwidth of 10 to 100 kHz. Then, deposit the coating (sample) onto the surface of the passive resonator. Next, identify the coupled vibration modes as local maxima in the monitored conductance spectrum and calculate the modal amplitude difference (AG) and the modal amplitude ratio (R). Finally, using the calculated AG and R parameters and their associated calibration curves, determine the surface mass density of the coating (sample).

[0032] In order to implement this measurement procedure, a device has been developed consisting of the following subsystems: - a mechanical and electronic interface where the substrate in which the CTSM coupled resonators have been implemented is inserted

[0033] - a microfluidic system that dispenses the sample onto the surface of the sensors and isolates the sample from the electrical contacts of the resonators;

[0034] - a signal source of a specific frequency, with high frequency stability and low phase noise, providing a signal between 1 and 300 MHz;

[0035] - a frequency synthesis subsystem;

[0036] - a signal conditioning and acquisition subsystem;

[0037] - a control subsystem based on the use of digital integrated circuits that allows coordinated control of the different subsystems that make up the device.

[0038] - a floating-point calculation subsystem that allows the implementation of the operations necessary to identify the resonant modes in the conductance spectrum and calculate the AG and R parameters.

[0039] Description of the drawings

[0040] For a better understanding of what is described in this document, some drawings are included which, as an example, represent a relationship of the figures of the proposed invention.

[0041] Figure 1 shows: (a) Top view of a 5 MHz CTSM resonator. (b) Bottom view of a 5 MHz CTSM resonator in SHORTED configuration, (c) Bottom view of a 5 MHz CTSM resonator in OPEN configuration, (d) Side view diagram of a CTSM resonator in OPEN configuration, (e) Side view diagram of a CTSM resonator in SHORTED configuration.

[0042] Figure 2 shows the experimental impedance spectrum of a 5 MHz CTSM resonator in OPEN configuration around its third harmonic (solid line) compared to the spectrum predicted by the finite element model (dots). (Bottom image) Experimental impedance spectrum of a 5 MHz CTSM resonator in SHORTED configuration around its third harmonic (line) compared to the spectrum predicted by the finite element model (dots).

[0043] Figure 3 shows the variation of the acoustic coupling coefficient K as a function of the distance between resonators. Experimental data for 5 MHz CTSM sensors in an OPEN configuration (circles), experimental data for 5 MHz CTSM sensors in a SHORTED configuration (diamonds), finite element simulation results (solid line), and the theoretical coupling predicted by Beaver's formula (dashed line) are shown. The experimental impedance spectrum of a 5 MHz CTSM resonator in a SHORTED configuration around its third harmonic (solid line) is compared to the spectrum predicted by the finite element model (dots). The simulated mode shapes for the sensor at series and parallel frequencies are inset.

[0044] Figure 4 shows the evolution of the third harmonic conductance spectrum of a 5MHz CTSM sensor in SHORTED configuration for successive gold depositions on one of its resonators. The arrows indicate increasing mass. (b) Evolution of the third harmonic resistance spectrum of a 5MHz CTSM sensor in OPEN configuration for successive gold depositions on one of its resonators. The arrows indicate increasing mass.

[0045] Figure 5 shows: (a) Modal amplitude difference as a function of the mass deposited on the active electrode measured in parts per million (ppm) of the reference resonance frequency for a CTSM sensor in SHORTED configuration with an inter-resonator distance (1.5 mm), (b) Ratio of coupled mode amplitudes as a function of the mass deposited on the active electrode measured in parts per million (ppm) of the reference resonance frequency for a CTSM sensor in SHORTED configuration with an inter-resonator distance (1.5 mm).

[0046] Description of a preferred embodiment

[0047] A preferred embodiment is cited as an example, independent of the object of the invention, the materials used in its manufacture, as well as the application methods and all accessory details that may arise, provided that they do not affect its essentiality.

[0048] With regard to Figures 1 to 5, the proposed device improves the stability and limit of detection (LoD) achieved by QCM sensor technology. Specifically, the proposed device consists of two quartz crystal resonators vibrating in shear mode. The resonators are semicircular and fabricated on a circular AT-cut quartz wafer 14 mm in diameter and 330 microns thick. One of these resonators (active) is electrically excited. The other (passive) is allowed to vibrate freely due to mechanical coupling. Thus, only two electrical contacts are needed to measure the CTSM sensor response. The resonators have two thin gold layers, approximately 100 nm thick, superimposed on the two faces of the quartz crystal that serve as electrodes. As shown in Figure 1(a), the upper electrode of the CTSM has the typical arrangement of a 5 MHz QCM with wrapped electrodes.

[0049] Two different CTSM sensor topologies have been manufactured: OPEN and SHORTED. These topologies differ only in the design of the lower electrodes. While in the OPEN configuration the passive electrode remains disconnected in an open circuit (Figure 1(c)), the lower passive electrode in SHORTED is short-circuited to the upper common electrode via a wrapped contact (Figure 1(b)). This difference significantly affects the sensor's response. Figure 1(d) shows the wiring diagram for a CTSM sensor in the OPEN configuration, while Figure 1(e) shows the diagram for the SHORTED configuration.

[0050] Figure 2 shows the experimental impedance spectrum measurement of a CTSM sensor with a fundamental frequency of 5 MHz in both OPEN (top panel) and SHORTED (bottom panel) configurations. The sensor response is evaluated around the third harmonic (15 MHz). Unlike traditional QCM sensors, where the impedance spectrum exhibits a minimum at series resonance and a maximum at parallel resonance, CTSM sensors exhibit two maxima and two minima. Figure 2 also includes the impedance spectrum simulated using a numerical model of the sensor based on the Finite Element Method (FEM) (points on the graph). By comparing the impedance spectrum predicted by the FEM model with the experimental results, it is possible to identify the different vibration modes of the CTSM sensors.In SHORTED sensors, the symmetric coupled mode appears at the first series resonance frequency (fs1), and the antisymmetric coupled mode appears at the second series resonance frequency (fs2). Mechanical energy is concentrated in the electrically excited (active) resonator at the first parallel resonance (fp1). However, energy is almost entirely transferred to the passive resonator at the second parallel resonance frequency (fp2). In contrast to the response of the SHORTED configuration, coupled modes appear at the parallel resonance frequencies (fp1 and fp2) in OPEN sensors. In both cases, the resonance frequency of the antisymmetric mode is higher than that of the symmetric mode. This behavior is entirely consistent with Spletzer's modal localization theory.

[0051] Figure 3 shows the coupling coefficient, K, experimentally measured on CTSM sensors with four different distances between resonators (0.5, 1, 1.5, 2, 2.5, and 3 mm). The coupling constant is calculated as K = (fs2 - fs1) ​​ / fs1 for SHORTED sensors and as K = (fp2 - fp1) / fp1 for OPEN sensors. Figure 3 includes the values ​​of K predicted by the FEM numerical model and those obtained using the analytical expressions proposed by Sykes and Beaver (WD Beaver, “Analysis of Elastically Coupled Piezoelectric Resonators,” J Acoust Soc Am, vol. 43, no. 5, pp. 972-981, 1968, doi: 10.1121 / 1.1910967). In all cases, an exponential decrease in K with distance is observed.

[0052] The OPEN and SHORTED sensors exhibit very similar K values ​​for small inter-resonator distances, specifically 0.5 mm and 1 mm. However, as the distance between resonators increases, the coupling level decreases more rapidly in the SHORTED topology than in the OPEN topology. This difference is not present in FEM simulations, which predict the same coupling for both topologies.

[0053] Good agreement was observed between the FEM numerical simulations and the analytical data, but some discrepancies were found. A slight overestimation of coupling was observed by the numerical model for small inter-resonator distances in the vicinity of 0.5 mm. Beaver's analytical solution fits the experimental K value perfectly for d = 0.5 mm, but shows a much faster decay of coupling with distance than the experimental observations and the FEM numerical model results. No significant differences were found in the responses of CTSM sensors fabricated with electrodes aligned along the crystallographic X or Z directions.

[0054] To evaluate the mass sensitivity of CTSM sensors, controlled amounts of gold were deposited by evaporation onto one of the resonators of the active CTSM sensor, while the other (passive) was left undisturbed. For these experiments, CTSM sensors with a resonator spacing of 1.5 mm were selected as a compromise between weak coupling and an electrode surface area large enough to facilitate gold deposition.

[0055] Figure 4 shows conductance spectra of successive mass depositions on a SHORTED sensor. Conductance was selected as the measured variable because the coupled modes in the SHORTED topology coincide with the impedance minima at fs1 and fs2 (see Figure 2(b)). At these frequencies, the conductance exhibits well-defined peaks with a good signal-to-noise ratio that are easy to track. The red arrows indicate the increasing direction of the mass deposited on the active electrode. As the mass increases, the peak conductance value of the symmetric coupled mode decreases, while it increases in the antisymmetric mode. A decrease in the resonant frequency is also observed in both modes, similar to what would be observed in a conventional QCM. An analogous response was observed in the OPEN topology (see Figure 3).In this case, the actual impedance value of the sensor was monitored (Figure 5), since its peaks correspond to the impedance maxima located at fp1 and fp2 and, therefore, to the coupled modes in this configuration. It should be noted that the noise present in the OPEN sensor measurements is higher. This is because the monitoring system used for characterization is specifically designed to work with the low impedances typical of the serial frequency.

[0056] Based on these results, two different metrics for analyzing mass sensitivity in weakly coupled resonators, reported in the literature (C. Zhao, GS Wood, J. Xie, H. Chang, SH Pu, and M. Kraft, “COMPARATIVE STUDY OF DIFFERENT OUTPUT METRICS FOR A THREE WEAKLY COUPLED RESONATOR SENSOR Nano Research Group, University of Southampton, UK MOE Key Laboratory of Micro and Nano Systems for Aerospace, Northwestern Polytechnical University of Southampton,” IEEE Transducers 2015, vol. 3, pp. 2196–2199, 2015), have been investigated: the difference in amplitudes between modes and their ratio. In the case of the SHORTED topology, the amplitude ratio is calculated as R = G1 / G2, where G1 and G2 represent the conductance values ​​at fs1 and fs2, respectively. The amplitude difference is calculated as AG = GI – G2. The equivalent metric can be calculated in the case of OPEN topology using the resistance value at fp1 and fp2.Using a numerical fitting algorithm, the peak conductance and resistance amplitudes were extracted from the experimental spectra. Figure 5 shows the different metrics calculated for a 5 MHz CTSM sensor in a shorted configuration versus the added mass density on the active resonator, measured in ppm at the reference resonant frequency. Both the amplitude ratio and the amplitude difference depend linearly on the difference in mass deposited on the surface of the two resonators, with a correlation coefficient r. 2very high (greater than 0.99 in both metrics). This result is consistent with the predictions of our numerical models and confirms that an acoustic wave sensor based on the modal localization principle has indeed been implemented. The sensitivity of these metrics can be derived from the slope of the lines in Figures 7 and 8. Although sensitivity is a fundamental parameter in a mass sensor, it is not suitable for comparing different sensor technologies. For this purpose, it is necessary to use the concept of Limit of Detection (LoD), defined by the International Union of Pure and Applied Chemistry (IUPAC) as the lowest concentration of an analyte that an analytical process can reliably detect (McNaught, AD; Wilkinson, A. IUPAC. Compendium of Chemical Terminology 2nd ed. (the “Gold Book”); Blackwell Scientific

[0057] Publications: Oxford, UK, 1997). This definition considers the influence of intrinsic sensor parameters such as sensitivity, baseline stability, and signal-to-noise ratio, but also other factors such as the sensitivity and specificity of the sensor coating. Since the contribution of sensor surface functionalization to the LoD depends on the application, an alternative definition is required to evaluate sensor performance. In this case, we use the instrument detection limit (IDL). For a gravimetric sensor, the IDL is defined as the minimum surface mass that can be detected. The IDL is defined as 3o / S, where o is the system noise and S is the sensitivity. The parameter o is measured as the standard deviation of the metric for ten measurements. The IDL for CTSM sensors using the modal amplitude variation (AG) metric is 0.47 ppm.The IDL for CTSM sensors using the modal amplitude ratio metric, R=GI / G2, is 0.58 ppm. The IDL for the resonance frequency variation of a QCM sensor is 1.46 ppm. Therefore, metrics based on CTSM sensors provide an IDL that is 2.5 to 3 times better than the traditional measurement of QCM sensors.

[0058] One of the additional advantages of modal localization in sensor development is the high common-mode rejection ratio (CMRR) of these devices. This high CMRR is directly related to the robustness against external interference that affects the coupled resonators of the proposed device.

[0059] To evaluate the CMRR of the CTSM sensors, the dependence of the amplitude difference on temperature is studied in the range of 23 °C to 27 °C. Measurements are taken without depositing mass on any of the resonators that make up the CTSM sensor. From the mass sensitivity obtained from Figures 7 and 8, the sensitivity of the modal amplitude difference, AG, to temperature is determined. A 4 °C difference produces a maximum variation of 0.0325 mS / °C in AG, which is equivalent to a change of 0.17 ppm / °C. In the case of the traditional resonant frequency measurement, a maximum variation with temperature of 0.5 ppm / °C is obtained. That is, the proposed CTSM sensor offers almost three times greater stability against temperature changes in air.

Claims

CLAIMS 1. Acoustic wave sensor based on the modal localization principle characterized by: - consist of two or more acoustic wave resonators made from the same AT-cut quartz substrate with a thickness on the order of 10 to 500 nanometers vibrating in shear mode, each of the resonators has two metal electrodes made by deposition with a thickness between 5 and 200 nanometers superimposed on the two faces of said quartz substrate; - exhibit a “weak coupling” of acoustic origin between the electromechanical responses of the aforementioned resonators; this “weak coupling”, which depends on the distance between the resonators, is manifested by a value of the coefficient K less than 0.09; the coefficient K is defined as K=(f2-fi) / fi, where fi is the resonance frequency corresponding to the maximum conductance (minimum impedance) of the first coupled mode and Í2 is the frequency corresponding to the maximum conductance (minimum impedance) of the second coupled mode; - the difference in amplitudes of their coupled modes (AG) and their ratio (R), defined as AG=GI-G2 and =GI / G2, where Gi is the maximum conductance (minimum impedance) of the first coupled mode and G2 is the maximum conductance (minimum impedance) of the second coupled mode, depend only on the difference in mass uniformly deposited on the surface of the resonators.

2. Acoustic wave sensor according to claim 1, wherein only one of the resonators or active resonator is electrically excited while the others or passive resonators are allowed to vibrate freely due to mechanical coupling, wherein the interconnection system required for its operation has only two electrical contacts.

3. An acoustic wave sensor according to claims 1 and 2, characterized in that the active resonator has an electrical impedance spectrum with as many minima around its resonant frequency as resonators have been fabricated in the quartz substrate, where each of these minima of impedance corresponds to a coupled mode of the complete resonator system 4. Acoustic wave sensor according to claim 1 and 2, wherein the electrodes of at least one passive resonator are connected by a short circuit.

5. Acoustic wave sensor according to claim 1 and 2, wherein the electrodes of at least one passive resonator are left disconnected in an open circuit.

6. Acoustic wave sensor according to claims 1 and 2, characterized in that the mass sensitivity of its resonant response, i.e., its instrument detection limit or IDL, is at least one order of magnitude greater than the mass sensitivity shown by the resonance frequency measurement of conventional quartz microbalance (QCM) sensors described by Sauerbrey.

7. Acoustic wave sensor according to claim 1 and 2, characterized by a common mode rejection ratio, CMRR, that is at least one order of magnitude better than that shown by the resonance frequency measurement of conventional quartz microbalance (QCM) sensors described by Sauerbrey.

8. A measurement method based on the modal localization principle comprising determining the surface mass density of a coating using the sensor described in claims 1-7, characterized by: a) Continuously monitoring, with a sampling period of between 0.01 and 1 second, the electrical impedance of the active resonator around its nominal resonant frequency with a bandwidth between 10 and 100 kHz. b) Depositing the coating (sample) onto the surface of the passive resonator. c) Identifying coupled vibration modes as local maxima in the conductance spectrum monitored in step a). d) Calculate the modal amplitude difference as AG=GI-G2, where Gi is the maximum conductance (minimum impedance) of the first coupled mode and G2 is the maximum conductance (minimum impedance) of the second coupled mode. e) Calculate the modal amplitude ratio, R=GI / G2, where G1 is the maximum conductance (minimum impedance) of the first coupled mode and G2 is the maximum conductance (minimum impedance) of the second coupled mode. f) From the parameters AG and R calculated in d) and e) and their associated calibration curves, determine the surface mass density of the coating (sample).

9. System implementing the procedure of claim 8, characterized by comprising: - a mechanical and electronic interface where the substrate in which the CTSM coupled resonators have been implemented is inserted - a microfluidic system that dispenses the sample onto the surface of the sensors and isolates the sample from the electrical contacts of the resonators; - a signal source of a specific frequency, with high frequency stability and low phase noise, providing a signal between 1 and 300 MHz; - a frequency synthesis subsystem; - a signal conditioning and acquisition subsystem; - a control subsystem based on the use of digital integrated circuits that allows coordinated control of the different subsystems that make up the device. - a floating-point calculation subsystem that allows the implementation of the operations necessary to identify the resonant modes in the conductance spectrum and calculate the AG and R parameters

Images