An apparatus for monitoring chemical reactions, and a system and a method thereof

Abstract

5 This invention provides an apparatus for monitoring chemical reactions, the apparatus comprising at least one layer adapted to excite surface phonon polariton, at least one layer adapted to excite localized surface plasmon polariton, and a substrate, wherein the layers are stacked alternately on the substrate, and are further adapted to exhibit a plasmon-10 phonon coupling mode upon exposure to radiation for vibrational fingerprinting of molecules from the monitored chemical reactions based on their refractive index. The invention further provides a system and method involving the apparatus. 1 for publication. 15

Description

AN APPARATUS FOR MONITORING CHEMICAL REACTIONS, AND A SYSTEM AND A METHOD THEREOFFIELD OF INVENTION

[0001] The invention relates to the art of chemical detection. Particularly, the invention relates to an apparatus for monitoring chemicals that performs refractive index-dependent vibration fingerprint retrieval.BACKGROUND OF THE INVENTION

[0002] Infrared (IR) spectroscopy refers to a method of using infrared light to detect the bonding between atoms in a molecule. This method is used in industries involved in material science to identify the chemical functional groups in a molecule. Light in the infrared range causes chemical bonds within a molecule to vibrate in different ways. The vibration of the chemical bonds is caused by the absorption of specific wavelengths of infrared light. The absorption of the wavelengths is specific to each bond, giving each functional group in a molecule a “fingerprint” for identification. During infrared spectroscopy, changes in the intensity of the reflected infrared light are investigated to determine the wavelengths of the absorbed light. With this, the vibrational fingerprint may be utilised to determine the functional groups and bonding in the sample molecule.

[0003] However, the use of infrared vibration fingerprint has several shortcomings. Mainly, the infrared vibrational fingerprints of some molecular' bonds may overlap with other infrared vibrational fingerprints. This issue may result in errors and ambiguities in experimental observations and will limit the use of the spectroscopy method in advanced applications.

[0004] Among the prior arts that relate to the systems and methods for infrared spectroscopy include US Patent No. 10,571 ,606B2, assigned to Trustees of Boston University, which discloses a plasmonic nanoantenna used to detect the infrared changes in molecules. The nanoantenna is utilised in infrared spectroscopy to detect the infrared vibrational signature of molecules. The prior art also discloses the method of fabrication of the plasmonic nanoantenna using nanostencil spectroscopy.ii05] Another prior ait is the work of Neubrech et. al, titled “Antenna Sensing of Surface Phonon Polaritons”. This prior art discloses an experiment related to the resonant coupling of surface plasmon polariton (SPP) and surface phonon polariton (SPhP) signals. In the experiment, gold nanowires are fabricated on a silicon dioxide ( SiO2) substrate by means of electron beam lithography. The gold and SiO2formed a coupling of the surface plasmon polariton vibration of gold and the surface phonon polariton vibration of SiCh. This prior art reports that the resonant coupling produces a strongly increased sensitivity in IR detection of the surface layers and suggests further developments in sensing applications.

[0006] Both of the aforementioned prior arts face the same problem of the overlapping of infrared vibrational fingerprints in molecules since the mode that is used to detect the molecular fingerprint is infrared. Accordingly, it is desirable to have a solution that enables differentiation of the overlapped infrared vibrational fingerprint signals.SUMMARY OF INVENTION

[0007] An objective of the present invention is to provide an apparatus, and a corresponding system and method, that enables the overlap of infrared vibrational fingerprints in conventional infrared spectroscopy to be differentiated. To achieve this, the invention enables the generation of a different molecular vibration fingerprint known as refractive index (Rl)-dependent vibrational fingerprint. The Rl-dependent vibrational fingerprint is used in conjunction with the conventional infrared vibrational fingerprint detection to differentiate the overlapped infrared vibrational fingerprint.

[0008] The invention intends to provide an apparatus for monitoring chemical reactions. In particular, the apparatus is a nanoantenna comprising a surface phonon polariton (SPhP) layer, a localized surface plasmon polariton (LSPP) layer, and a substrate. The layers are adapted to exhibit a plasmon-phonon coupling mode upon exposure to radiation for vibrational fingerprinting of molecules.

[0009] Preferably, regarding the apparatus, the layer that is adapted to excite localized surface plasmon polariton is selected from a group of electrically conductive elements that include gold.

[0010] Preferably, regarding the apparatus, the layer that is adapted to excite surface phonon polaritons is selected from a group of polar dielectric materials that include silicon dioxide.

[0011] Preferably, regarding the apparatus, any one of the layers is attached to the substrate, and the substrate comprises material selected from a group of fluoride compounds that include barium fluoride.

[0012] Preferably, regarding the apparatus, the layers of the apparatus are arranged such that the first layer is an SPhP layer, the second layer is an LSPP layer, the third layer is an SPhP layer and the fourth layer is an LSPP layer. There is no restriction on the number of layers; any stacked arrangement of SPhP and LSPP layers is viable, and it is not confined to just four layers.

[0013] Preferably, regarding the apparatus, it has a height between substantially 50 nm to substantially 500 nm.

[0014] The invention further intends to provide a system for monitoring chemical reactions comprising the apparatus, a fluid chamber for receiving chemical reactants, and a radiation source that emits radiation towards the apparatus. The layers of the apparatus are stacked alternately on the substrate. They are further adapted to exhibit a plasmon-phonon coupling mode upon exposure to radiation received from the radiation source to excite the reactants present within the fluid chamber for vibrational fingerprinting of molecules of the reactants undergoing chemical reactions based on their refractive index.

[0015] Preferably, the system further comprises a radiation detector that receives radiation reflected from the apparatus for converting them into detection data, and a computer interfaced with the radiation detector that operates one or more modules.

[0016] Preferably, regarding the system, the computer operates a principal component analysis module, which processes the detection data by performing dimensional reduction thereupon.

[0017] Preferably, regarding the system, the computer further operates an identification module that receives the processed detection data as input and provides identified molecules within the chemical reactants as output. In particular, the identification of molecules is based on classification of vibration signals within their chemical bonds.

[0018] Preferably, regarding the system, the apparatus is disposed within the fluid chamber for its layers to be in substantial contact with the chemical reactants received by the fluid chamber.

[0019] Preferably, regarding the system, the apparatus is substantially disposed below the radiation source.

[0020] Preferably, regarding the system, the fluid chamber further comprises an inlet and an outlet, for the chemical reactants to be received by the fluid chamber and for the chemical reactants to be directed away from the fluid chamber, respectively.

[0021] Preferably, regarding the system, wherein the layer of the apparatus that is adapted to excite localized surface plasmon polariton is selected from a group of electrically conductive elements that include gold.

[0022] Preferably, regarding the system, the layer of the apparatus that is adapted to excite surface phonon polaritons is selected from a group of polar dielectric materials that include silicon dioxide.

[0023] Preferably, regarding the system, any one of the layers of the apparatus is attached to the substrate, and the substrate comprises material selected from a group of fluoride compounds that include barium fluoride.

[0024] Preferably, regarding the system, the layers of the apparatus are arranged such that the first layer is an SPhP layer, the second layer is an LSPP layer, the third layer is an SPhP layer and the fourth layer is an LSPP layer. The stacked arrangement of SPhP and LSPP layers is not confuted to just four layers.

[0025] The invention also intends to provide a method for monitoring chemical reactions, comprising the steps of preparing an apparatus for monitoring chemical reactions that comprises at least one layer that excites localized surface plasmon polariton, at least one layer that excites surface phonon polaritons, and a substrate, preparing a fluid chamber having chemical reactants therewithin, and emitting radiation towards the apparatus, by a radiation source. The layers of the apparatus are stacked alternately on the substrate, and are further adapted to exhibit a plasmon-phonon coupling mode upon exposure to radiation received from the radiation source to excite the reactants present within the fluid chamber for vibrational fingerprinting of molecules of the reactants undergoing chemical reactions based on their refractive index.

[0026] Preferably, the method further comprises steps of receiving radiation reflected from the nanoantenna for conversion into detection data by a radiation detector, and providing the detection data, by the radiation detector, to a computer that operates one or more modules.

[0027] Preferably, the method further comprises the steps of processing the detection data by performing dimensional reduction thereupon, by a principal component analysis module, receiving the processed detection data as input, by an identification module; and providing identified molecules within the chemical reactants as output, by the identification module. In particular, the identification of molecules as performed by the identification module is based on classification of vibration signals within their chemical bonds.

[0028] One skilled in the art will readily appreciate that the present invention is well adapted to carry out the objects and obtain the advantages mentioned, as well as those inherent therein. The embodiments described herein are not intended as limitations on the scope of the present invention.BRIEF DESCRIPTION OF THE DRAWINGS

[0029] To facilitate an understanding of the present invention, there are illustrated in the accompanying drawings the preferred embodiments, from an inspection of which when considered in connection with the following description, the invention, its construction, and operation and many of its advantages would be readily understood and appreciated.

[0030] FIG. 1 illustrates an apparatus according to a first embodiment of the present invention.

[0031] FIG. 2 shows the scanning electron microscope (SEM) images, atomic force microscope (AFM) images, and energy dispersive x-ray (EDX) mappings of the apparatus.

[0032] FIG. 3 illustrates an apparatus according to a second embodiment of the present invention.

[0033] FIG. 4 are graphical representations of dispersion curves of surface plasmon polariton (SPP) in the upper panel, and the local surface plasmon polariton (LSPP)-surface phonon polariton (SPhP) coupling system in the lower panel that made up the above apparatuses.

[0034] FIG. 5 shows the simulation results of the apparatus as an SP-PhP platform for molecules with varying Δn, investigating the relationship between Δn and AZ. (a) The simulated reflection mapping. Rupperand Rlowerare upper / lower polaritonic resonances, (b) The permittivity of SiO2. (c) The Δn-ΔAI transition curve was obtained by calculating the intensity integral of the SPhP vibration within longitudinal ( ωLO) and transverse optic (ωTO) phonon frequencies.

[0035] FIG. 6 illustrates a system in which the above apparatuses may be integrated thereinto to cany out the monitoring of chemical reactions.

[0036] FIG. 7 illustrates a drawing of a first example experiment to monitor glucose enzymatic reaction (GER).

[0037] FIG. 8 illustrates real-time three-dimensional plots obtained from the first example experiment, which includes differential absorbance spectra versus wavenumber and concentration when using the apparatus to detect one analyte at a time and their corresponding regression curves, for glucono-l,5-lactone, glucose, hydrogen peroxide H2O2, and glucose oxidase (GOD), respectively.

[0038] FIG. 9 illustrates the spectra obtained from the first example experiment, and the schematic illustration of the monitored enzymatic reaction path.

[0039] FIG. 10 illustrates the real-time weights of all the reactants and products in the experiment obtained using a conventional nanorod-based platform.

[0040] FIG. 11 illustrates the structures of severe acute respiratory syndrome coronavirus 1 (SARS-CoV) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus.

[0041] FIG. 12 illustrates the collected Transverse Magnetic (TM)-polarised phonon signals of the present invention as a nanoantenna system for the different concentration configurations of SARS-CoV and SARS-CoV-2.

[0042] FIG. 13 illustrates the collected Transverse Electric (TE)-polarised phonon signals of the present invention as a nanoantenna system for the different concentration configurations of SARS-CoV and SARS-CoV-2.

[0043] FIG. 14 illustrates the results of dimensionality reduction of TE-polarised vibration and TE-polarised vibration combined spectra.

[0044] FIG. 15 illustrates the support vector machine (SVM) outcome of the data obtained from the principal component analysis.

[0045] FIG. 16 illustrates the support vector machine (SVM) outcome of the data obtained from principal component analysis from a similar procedure that is carried out using a conventional plasmonic nanoantenna system.

[0046] FIG. 17 illustrates machine learning-based concentration prediction of SARS-CoV and SARS-CoV-2 strains.

[0047] FIG. 18 illustrates the steps to perform hyperspectral imaging of coronavirus.

[0048] FIG. 19 illustrates a schematic representation of the distribution of samples monolayers that are anchored onto the above apparatus.

[0049] FIG. 20 illustrates the data cubes of the acquired 460,800 spectra of TM- polarisation and TE-polarisation hyperspectral images.

[0050] FIG. 21 illustrates the resulting protein monolayer distribution in the prediction results.

[0051] FIG. 22 illustrates the fabrication process of the apparatus whereby its layers are built up on a substrate.

[0052] FIG. 23 and FIG. 24 illustrate a comparison of electric field distribution between the apparatus of the present invention and the conventional mono-layer antenna.

[0053] FIG. 25 shows a comparison of the polarisation range between the apparatus of the present invention and a conventional mono-layer nanorod.

[0054] FIG. 26 shows a comparison of the incident range between the apparatus of the present invention and a conventional mono-layer nanorod.DETAILED DESCRIPTION OF THE INVENTION

[0055] The present invention relates to an apparatus, a system and a method for monitoring chemical reactions to retrieve molecular fingerprints of chemical reactants. According to the concept of the invention, the apparatus is more specifically a form of nanoantenna device that detects surface phonon polariton (SPhP) vibration signals and surface plasmon polariton (SPP) vibration signals. The two vibration signals couple together to form a refractive index (RI) vibration fingerprint of the target molecule. The RI vibration fingerprint obtained is used to differentiate overlapped infrared vibration fingerprint, thereby achieving higher sensitivity and accuracy of detection.

[0056] For context, the Rl-dependent vibration comprises a specific type of surface phonon polariton (SPhP) oscillation, formed when polar optical phonons interact with long- wavelength fields, i.e. infrared light.

[0057] The invention will now be described in greater detail, by way of example, with reference to the figures.

[0058] FIG. 1 illustrates the apparatus 1, which comprises at least one or an array of layered structures 11 of a first embodiment that is in the shape of a har. Each layered structure 11 is to be considered as a nanoantenna. Furthermore, each layered structure 11 comprises two polar dielectric crystal layers 11a and two metal layers 11b. In particular, each layer within the layered structure 11 is stacked on top of each other in an alternating manner on a substrate 13 that is transparent. Preferably, the polar dielectric crystal layers Ila are adapted to excite surface phonon polariton vibrations, while the metal layers 11b are adapted to excite localized surface plasmon polariton vibrations.

[0059] FIG. 2 illustrates the scanning electron microscope (SEM) images, atomic force microscope (AFM) images, and energy dispersive x-ray (EDX) mappings of the first embodiment of the apparatus 1. According to FIG. 2, the polar dielectric crystal layer Ila comprises silicon dioxide (SiO2), while the metal layer 11b comprises gold (Au). Furthermore, the substrate 13 comprises barium fluoride (BaF2). In addition to this, the height of the apparatus 1 is substantially within the range of 50nm to 500nm.

[0060] FIG. 3 illustrates the apparatus 1, which comprises at least one or an array of layer structures 12 of a second embodiment that is in the shape of a cross. Likewise, each layered structure 12 is to be considered as a nanoantenna. Furthermore, each layered structure 12 comprises one polar dielectric crystal layer 12a and one metal layer 12b, with the metal layer 12b stacked on top of the polar dielectric crystal layer 12a, and with the polar dielectric crystal layer 12a being on top of the substrate 13 that is transparent. Similarly, the polar dielectric crystal layer 12a is adapted to excite surface phonon polariton vibrations, while the metal layer 12b is adapted to excite local surface plasmon polariton vibrations. It is to be noted that the composition and dimensions of the layered structure 12of the second embodiment may be similar to as previously described for the layered structure 11 of the first embodiment.

[0061] It is to be noted that as layered structure 11 of the first embodiment is the preferred embodiment, all mentions of a “layered structure” shall be referred thereto. However, its descriptions thereof may similarly be applicable to the layered structure 12 of the second embodiment.

[0062] From the schematic illustration of the embodiments as shown in the figures so far, it is shown that the arrangement of the layered structure 11 is in such a way that the metal layer lib is directly coupled to the polar dielectric crystal layer Ila. This coupling (Δn) causes the plasmon on the surface of the nanoantenna to have a resonance frequency shift from a first frequency (cm) to a second frequency (®B). This resonance frequency shift is as shown in the dispersion curve of the LSPP-SPhP hybrid of FIG. 4 (upper panel), whereby different positions of the split imply varying plasmonic resonance frequencies. This change from A+A- to B+B- indicates a change in resonance frequency. As the SPhP Reststrahlen band remains fixed, altering the plasmonic frequency (A®) yields distinct coupling strengths, consequently influencing the SPhP vibration intensity (AT) as shown in FIG. 4 (lower panel). The relationship between \ / t and AZ indicates that the apparatus 1 behaves similarly to a SP-PhP nanoantenna system that can leverage the vibrational variations of SPhP within the Reststrahlen band to provide molecular refractive index feature Δn.

[0063] The Rl-dependent vibration occurs in the Reststrahlen band of polar dielectric crystals, situated between the longitudinal optic (LO) and transverse optic (TO) phonon frequencies of polar dielectric crystal layer Ila, exhibiting notable sensitivity to the molecular refractive index.

[0064] FIG. 5 illustrates the Finite-difference time-domain (FDTD) simulation results of the Rl-dependent vibrations. The panel labelled (a) indicates the simulated reflection mapping of phonon intensity (AT) as a function of refractive index (Δn). The panel labelled (b) indicates the permittivity of the phonon layer, while the panel labelled (c) indicates the\n-\I transition curve obtained by calculating the intensity integral of the SPhP vibration within longitudinal (®LO) and transverse optic (®ro) phonon frequencies.

[0065] The apparatus is shown in FIG. 1 and FIG. 3 may be integrated into a system for monitoring chemical reactions as shown in FIG. 6. The system comprises the apparatus 1, a fluid chamber 2 having an inlet 3 and an outlet 4, a radiation source 5, a radiation detector 6, and a computer 7.

[0066] Preferably, the fluid chamber 2 is a microfluidic chamber that receives chemical reactants. More specifically, chemical reactants are to be received by the fluid chamber 2 via its inlet 3, and are to be directed away from the fluid chamber 2 via its outlet 4.

[0067] Preferably, the radiation source 5 emits electromagnetic radiation in the infrared spectrum, and the radiation detector 6 is provided to detect electromagnetic radiation in the infrared spectrum.

[0068] Preferably, the apparatus 1 is disposed within the fluid chamber 2, more specifically, at the walls of the fluid chamber 2, for its layered structures 11 to be in substantial contact with the chemical reactants received by the fluid chamber 2. Furthermore, the apparatus 1 is substantially disposed below the radiation source 5. Not only that the walls of the fluid chamber 2 are transparent as well.

[0069] Preferably, computer 7 is connected to the radiation detector 6. In particular, the computer 7 further comprises a processing unit 7a for processing detection data from the radiation data 6, and a memory unit 7b for storing the detection data and processed detection data.

[0070] ft is further shown in FIG. 6 that processing unit 7a operates modules that may include a principal component analysis (PCA) module, an identification module and a machine learning module. It should be noted that the presented modules need not be in a software embodiment, and may be a hardware embodiment where they are connected to the processing unit 7a or they are their own independent computer system. Ancillary modules may be included to provide support for the aforementioned modules.

[0071] In particular, the PCA module processes the detection data from the radiation detector 6 by performing dimensional reduction thereupon. More specifically, the detection data are the molecular vibration signals and RI-RI-dependent vibration signals of the chemical reactants.

[0072] In particular, the identification module receives the processed detection data from the PCA module as input, and shall provide identified molecules within the chemical reactants as output. The identification of molecules is based on the classification of vibration signals within their chemical bonds.

[0073] The machine learning module is interfaced with the identification module for facilitating the aforementioned classification. In particular, the machine learning module deploys a trained machine learning model that was trained by a library of Rl-dependent fingerprint spectra and infrared spectra obtained through single analyte experiments. The machine learning module may also be configured to train an existing machine learning model to identify the fingerprints for future single-analyte and multiple-analyte experiments. The data of the machine learning model, such as its weights and biases, is stored in memory unit 7b.

[0074] The system as described can be used in several use cases such as monitoring chemical reactions, more specifically enzymatic reactions, classifying viral molecules, predicting viral molecules in a solution, and hyperspectral imaging of viral molecules.

[0075] With this, a method for monitoring chemical reactions may now be described. It is noted that the steps described for the method are not to be interpreted as limiting, and minor modifications to the steps (e.g., combination, additions, omissions, or swaps) are permissible by a skilled person without substantial deviation from as described. Furthermore, all steps or some of the steps may occur in a simultaneous or nonsimultaneous manner.

[0076] Generally, the method may begin with a first step that involves pre-treating the apparatus 1 for a catalyst of a chemical reaction to be immobilised on its layered structures 11.

[0077] Next, there is a second step that involves integrating the apparatus 1 into the fluid chamber 2 such so that its structured layers 11 arc capable of being in contact with chemical reactants that are target sample(s).

[0078] Next, there is a third step that involves injecting the sample(s) into the inlet 3 of the fluid chamber 2 such that the chemical reactants are transferred into an observation area to come in contact with the structured layers 11 of the apparatus 1.

[0079] Next, there is a fourth step that involves emitting infrared light onto the chemical reactants and the apparatus 1 to obtain the reflected infrared light by the radiation detector 6, which is preferably located in the vicinity of the reflected infrared light. With this, the radiation detector 6 obtains detection data.

[0080] Finally, there is a final step that involves transferring the obtained detection data to the computer 7 for analysis.

[0081] Tt is to be noted that the method for monitoring chemical reactions may further involve the step of collecting Transverse Magnetic (TM)-polarisation vibration signatures and Transverse Electric (TE)-polarisation vibration signatures of the sample using the infrared light. The signatures collected are processed by the PCA module and the identification module of the computer 7.

[0082] It is to be noted that the method for monitoring chemical reactions may further involve the steps of collecting the TM-polarisation vibration signatures and TE -polarisation vibration signatures by the radiation detector 6, and performing dimensionality reduction thereupon to obtain processed data, feeding the processed data to the machine learning model for it to train a machine learning model, and testing the machine learning model for validating accuracy of its intended function.

[0083] It is to be noted that the method for monitoring chemical reactions may further involve the steps of making predictions in the classification of viral molecules, making predictions in the concentration of viral solutions, and hyperspectral imaging of viral molecules.

[0084] From hereon, evaluations that were carried out to validate the performance of the apparatus, system, and method of the present invention will be described. It is to be noted that parameters defined or determined in the evaluations are not meant to be interpreted as limitations to the scope of the invention.

[0085] FIG. 7 illustrates a schematic drawing of a first example experiment to monitor glucose enzymatic reaction (GER).

[0086] The detailed steps of preparing the apparatus 1 for the first example experiment are as follows. It is noted that the steps described for the method are not to be interpreted as limiting, and minor modifications to the steps (e.g., combination, additions, omissions, or swaps) are permissible by a skilled person without substantial deviation from as described. Furthermore, all steps or some of the steps may occur in a simultaneous or nonsimultaneous manner.

[0087] In a first step, the apparatus 1, more specifically its layered structures, is immersed in acetone solution at room temperature for substantially thirty minutes.

[0088] Next, in a second step, the apparatus is washed about three times with deionised water and isopropyl alcohol (1PA), and subsequently dried using nitrogen.

[0089] Next, in a third step, the surface of the apparatus 1 is treated with oxygen plasma to generate hydroxyl groups.

[0090] Next, in a fourth step, the apparatus 1 is immersed in an ultrasonically pretreated graphene oxide (GO) dispersion that has a concentration of substantially 1 mg / mf at 50 °C to immobilize GO onto the surface of the apparatus 1 via covalent bonding.

[0091] Next, in a fifth step, the apparatus 1 is incubated in a cross-linker solution of 20inM of l-ethyl-3-(3-dimethiylaminopropyl) carbodiimide (EDC) and 40mM N-hydroxy succinimide (NHS) for substantially sixty minutes to enable chemical Unking of GO with the catalyst glucose oxidase (GOD). The pH is adjusted to 6.1 using a 2-(N-Morpholino) ethanesulfonic acid (MES) buffer that has pH 5.5.

[0092] Next, in a sixth step, the heated chip is immersed into the coupling buffer (Img / mL GOD in phosphate-buffered saline (PBS) buffer) for substantially 90 minutes, where pH is set to substantially 7.2 by the PBS buffer.

[0093] Finally, the apparatus 1, having been functionalised, is thoroughly washed with PBS buffer and deionised water and dried at 4 °C. Successful immobilisation of the GOD onto the surface of apparatus 1 may be confirmed by the observation of amide vibrations of GOD in the measured spectra.

[0094] Before initiation of the first example experiment, the apparatus 1 is integrated onto the system in which its layered structures are exposed to the sample in the microfluidic chamber 2.

[0095] The detailed steps of the first example experiment are as follows. It is noted that the steps described for the method are not to be interpreted as limiting, and minor modifications to the steps (e.g., combination, additions, omissions, or swaps) are permissible by a skilled person without substantial deviation from as described. Furthermore, all steps or some of the steps may occur in a simultaneous or nonsimultaneous manner.

[0096] In a first step, about 110 mM glucose solution is injected into the fluid chamber 2 to initiate the experiment.

[0097] In a second step, the radiation source 5 emits infrared light into the fluid chamber 2 through the apparatus 1, and the reflected infrared light is obtained by the radiation detector 6 as detection data. The detection data is transferred to the computer 7 for analysis. A series of time-varying reflectance spectra is collected at substantially one-minuteintervals. Notably, environmental disturbances are eliminated by measuring the background spectrum before each spectrum collection.

[0098] In a final step, the contents of the fluid chamber 2 arc emptied out via the outlet 5.

[0099] FIG. 8 illustrates real-time 3D plots of differential absorbance spectra versus wavenumber and concentration when using the apparatus 1 to detect one analyte at a time and their corresponding regression curves, for glucono- 1,5-lactone, glucose, hydrogen peroxide (H2O2), and GOD.

[0100] Whereas, FIG. 9 shows the spectra obtained from the example experiment as well as the schematic illustration of the monitored enzymatic reaction path. The spectra may be provided to the machine learning module.

[0101] Noticeably, the weights of all the reactants and products in the first example experiment can be distinguished in real-time. Particularly, line (ii) and line (iv) representing amide bonds in GOD and O-H bonds H2O2show a distinct difference, where line (iv) increases with time, and line (ii) remains constant throughout the experiment. The difference mentioned cannot be achieved by conventional nanorod-based platforms that arc unable to acquire Rl-dependent fingerprints.

[0102] For comparison, FIG. 10 shows the real-time weights of all the reactants and products in the experiment obtained using a conventional nanorod based platform, line (ii) and line (iv) show significant overlapping with each other as the representation of the lines are incorrect. Line (ii) does not increase with time as it represents the catalyst of the reaction. It is shown that the nanorod based platform cannot differentiate the O-H bonds of H2O2from the amide groups in the GOD.

[0103] The apparatus, system and method of the present invention can be used to classify SARS-CoV and SARS-CoV-2 accurately.

[0104] FIG. 11 shows a schematic illustration of the structures of SARS-CoV and S ARS- CoV-2 virus.

[0105] The detailed steps to classify SARS-CoV and SARS-CoV-2 are as follows. It is noted that the steps described for the method are not to be interpreted as limiting, and minor modifications to the steps (e.g., combination, additions, omissions, or swaps) are permissible by a skilled person without substantial deviation from as described. Furthermore, all steps or some of the steps may occur in a simultaneous or nonsimultaneous manner.

[0106] In a first step, SARS-CoV and SARS-CoV-2 with varying concentrations were loaded onto the system.

[0107] Next, in a second step, the step of collecting the Transverse Magnetic (TM)- polarised infrared vibrational signals and the Transverse Electric (TE)-polarised phonon signals is performed. FIG. 12 and FIG. 13 show the results of this collection, where the FIG. 12 illustrates the collected TM-polarised infrared vibrational signals, and the FIG. 13 illustrates the collected TE-polarised phonon signals of nanoantenna system for the different concentration configurations of SARS-CoV and SARS-CoV-2.

[0108] Next, in a third step, the results are fed into the PCA module for dimensional reduction to obtain data as shown in FIG. 14. As seen in the data, most data points exhibit clear separation. Crucially, upon individual analysis of the TM-polarised signal or TE- polarised signal using a PCA algorithm, effective separation of these overlapping spectra was not achieved. This suggests that the molecular features contained in TM- and TE- polarised signals are crucial and different, and their combination enhances the decoupling of overlapping spectra by the PCA algorithm.

[0109] Finally, the processed data from the PCA module is further fed into an identification module for species identification. The confusion map of the support vector machine (SVM) outcome is as shown in FIG. 15, which indicates a classification accuracy of 93.4% for SARS-CoV and SARS-CoV-2.

[0110] For comparison, a similar procedure was carried out using a conventional plasmonic nanoantenna system. Subsequently, these signals were classified using PCA andS VM algorithms. According to the results shown in FIG. 16, the PCA data points associated with this approach exhibited significant overlap, and the classification accuracy of SVM was only 22.8%.

[0111] FIG. 17 shows steps that are related to machine learning-based concentration prediction of SARS-CoV and SARS-CoV-2 strains. It is noted that the steps described arc not to be interpreted as limiting, and minor modifications to the steps (e.g., combination, additions, omissions, or swaps) are permissible by a skilled person without substantial deviation from as described. Furthermore, all steps or some of the steps may occur in a simultaneous or non-simultaneous manner.

[0112] In a first step, the TM-polarisation spectra and TE-polarisation spectra of SARS- CoV, SARS-CoV-2 and SARS-CoV / SARS-CoV-2 mixtures are fed into the identification module.

[0113] Next, in a second step, newly measured experiment datasets with different concentrations ranging from substantially 0 to 400 ng / pL of SARS-CoV and SARS-CoV- 2, as well as mixtures of both, are provided into the machine learning module for the prediction of concentration training using a deep neural network (DNN) model.

[0114] Finally, in a third step, the machine learning model provides the identification module with the trained DNN T model, and it was found that the DNN model was able to predict the concentration of SARS-CoV and SARS-CoV-2 with high accuracy, exhibiting an average error of substantially 8.08%.

[0115] FIG. 18 illustrates the steps for the invention to perform hypcrspcctral imaging of coronavirus from a heterogeneous sample. It is noted that the steps described for the method arc not to be interpreted as limiting, and minor modifications to the steps (e.g., combination, additions, omissions, or swaps) are permissible by a skilled person without substantial deviation from as described. Furthermore, all steps or some of the steps may occur in a simultaneous or non-simultaneous manner.

[0116] The first step involves immobilizing a monolayer of heterogeneously mixed proteins onto the system. This step is promptly labelled (a) in FIG. 18.

[0117] Next, there is a second step that involves hyperspectral imaging under long-arm polarisation, where the plasmon mode of the nanoantenna is excited to detect the extinction coefficient (k) feature of the heterogeneous sample. This step is promptly labelled (b) in FIG. 18.

[0118] Next, there is a third step that involves hyperspectral imaging under short-arm polarisation, where the plasmon-phonon coupling mode is excited to acquire the refractive index (n) feature of the heterogeneous sample. This step is promptly labelled (c) in FIG.18.

[0119] Next, there is a fourth step that involves training and validating a deep neural network model with the data acquired previously in the second step and the third step. This step is promptly labelled (d) in FIG. 18.

[0120] The fifth and final step involves having the trained model identify 460,800 overlapping spectra in hyperspectral images each having 480 pixels by 480 pixels, for predicting the component distributions for the heterogeneous sample. This step is promptly labelled (d) in FIG. 18.

[0121] FIG. 19 to FIG. 21 illustrate the experimental data of hyperspectral imaging of coronavirus monolayers. Tn particular, FIG. 19 illustrates the distribution of samples monolayers that are anchored onto the layered structure 11. In particular, FIG. 20 illustrates the acquired 460,800 spectra of TM-polarisation and TE-polarisation hyperspectral images, which are then fed into the DNN model for training. As a result, the DNN model is then able to predict the distribution of SARS-CoV-2 and bovine serum albumin (BSA), with 100% training accuracy and 93% validation accuracy. In particular, FIG. 21 illustrates that the resulting protein monolayer distribution in the prediction aligns with the observed immobilisation of the sample.

[0122] FIG. 22 illustrates the fabrication process of the one or an array of layered structure 11 on the substrate 13. The layered structure 11 are preferably nanofabricated using electron beam lithography. It is noted that the steps described for the process are not to be interpreted as limiting, and minor modifications to the steps (e.g., combination, additions, omissions, or swaps) are permissible by a skilled person without substantial deviation from as described. Furthermore, all steps or some of the steps may occur in a simultaneous or non-simultaneous manner.

[0123] In a first step, the substrate 13 is cleaned using acetone by undergoing ultrasonic treatment for substantially 10 minutes.

[0124] Next, there is a second step that involves rinsing the substrate 13 in isopropanol and drying it with nitrogen.

[0125] Next, there is a third step that involves performing oxygen plasma treatment onto the substrate 13 for substantially 5 minutes.

[0126] Next, there is a fourth step that involves spin-coating of a layer of polymethyl methacrylate (PMMA) electron beam lithography resist (950 PMMA A5) at 300 rpm, with the layer being substantially 400 nm thick.

[0127] Next, there is a fifth step that involves thermal baking the substrate 13 and spin coating of a commercial electron-conducting polymer such as Espacer™ 300Z at a speed of 2000 rpm to eliminate charge accumulation during electron beam exposure.

[0128] Next, there is a sixth step that involves exposing a pattern, which corresponds to the intended distribution of the layered structure 11, onto the substrate 13 using electron beam lithography.

[0129] Next, there is a seventh step that involves developing the substrate 13 with deionised water, an MIB / IPA mixture having a ratio of about 1 :3, and isopropanol.

[0130] Next, there is an eighth step that involves sequentially depositing one or more layers with a desired thickness using electron beam evaporation. The sequence of the deposited layers, from bottom to top (i.e. in a normal direction from the surface of substrate 13), may be SiC>2 > Ti > Au > Ti > Si()> > Ti > Au, wherein the Si()> layers are the adapted to excite surface phonon polariton vibrations, the Au layers are adapted to excite local surface plasmon polariton vibrations, and the Ti layers are intermediary layers between the said layers.

[0131] Finally, there is a ninth step that involves immersing the apparatus 1 in acetone for 24 hours to remove the unexposed resist and obtain a final pattern of layered structure 11 on the apparatus 1.

[0132] FIG. 23 and FIG. 24 illustrate a comparison of electric field distribution between the layered structure 11 of the apparatus 1 in the present invention and conventional mono- layer antenna.

[0133] The left panel of FIG. 23 illustrates the electric field distribution of a conventional nanorod antenna while the right panel of FIG.23 illustrates the spatial profile of the electric field intensity of the nanorod antenna.

[0134] The left panel of FIG. 24 illustrates the electric field distribution of the layered structure 11 of the apparatus 1 of the present invention while the right panel of FIG. 24 illustrates the spatial profile of the electric field intensity of the layered structure 11 of the apparatus 1 of the present invention.

[0135] It is revealed that the layered structure 11 of the apparatus 1 of the present invention exhibits enhanced sensitivity in comparison to its conventional single-layer nanoantenna. The simulation shows that the electric field exposure rate for the single-layer nanoantenna in free space is 52.7%, whereas the double-layer nanoantenna as per the present invention can achieve an exposure rate of 72.3%. The higher exposure rate implies increased molecular entry into the near- field of the antenna, thereby enhancing molecular spatial overlap with the electric field and enabling more sensitive detection.

[0136] FIG. 25 illustrates a comparison of the polarisation range of the layered structure 11 of the apparatus 1 in the present invention and a conventional mono-layer nanorod. The top panels in FIG. 25 are schematic illustrations of the layered structure 11 of the apparatus 1 (top left panel) and the conventional nanorod (top right panel), while the bottom panels are corresponding polarisation angle against spectrum wavenumber mapping of the of the layered structure 11 of the apparatus 1 (bottom left panel) and the conventional nanorod (bottom right panel).

[0137] As shown in FIG. 25, there is no significant difference observed in the polarisation response between the layered structure 11 of the apparatus 1 and the conventional mono-layer nanorod. Both the layered structure 11 of the apparatus 1 and the single-layer nanorods exhibit a polarisation variation range of 35°, within which the resonance strength of the antenna exceeds half of that observed when the polarisation angle is 0.

[0138] FIG. 26 illustrates a comparison of the incident range of the layered structure 11 of the apparatus 1 in the invention and a conventional mono-layer nanorod. The top panels in FIG. 26 are schematic illustrations of the layered structure 11 of the apparatus 1 (top left panel) and the conventional nanorod (top right panel), while the bottom panels are the incident angle against spectrum wavenumber mapping of the layered structure 11 of the apparatus 1 (bottom left panel) and the conventional nanorod (bottom right panel).

[0139] As shown in FIG. 26, the angle of incidence range of the layered structure 11 of the apparatus 1 also does not fall short in the angle of incidence range of the conventional mono-layer nanorod. Both the layered structure 11 of the apparatus 1 and the conventional mono-layer nanorod have a 50° variation range, where the resonance strength is greater than half at a normal angle of incidence.

[0140] With this, the details pertaining to an apparatus for monitoring chemical reactions, and its corresponding system and method, have been described.

[0141] The present disclosure includes as contained in the appended claims, as well as that of the foregoing description. Although this invention has been described in its preferredform with a degree of particularity, it is understood that the present disclosure of the preferred form has been made only by way of examples and that numerous changes in the details of construction and the combination and arrangements of parts may be resorted to without departing from the scope of the present invention.

Claims

CLAIMS:

1. An apparatus for monitoring chemical reactions, the apparatus comprising: at least one layer adapted to excite surface phonon polariton; at least one layer adapted to excite localized surface plasmon polariton; and a substrate; wherein the layers are stacked alternately on the substrate, and are further adapted to exhibit a plasmon-phonon coupling mode upon exposure to radiation for vibrational fingerprinting of molecules from the monitored chemical reactions based on their refractive index.

2. The apparatus according to claim 1, wherein the layer that is adapted to excite localized surface plasmon polariton is selected from a group of electrically conductive elements that include gold.

3. The apparatus according to claim 1 or 2, wherein the layer that is adapted to excite surface phonon polaritons is selected from a group of polar dielectric materials that include silicon dioxide.

4. The apparatus according to any one of the preceding claims, wherein any one of the layers is attached to the substrate, and the substrate comprises material selected from a group of fluoride compounds that include barium fluoride.

5. The apparatus according to any one of the preceding claims, wherein its layers sequentially comprise: a first layer that excites surface phonon polariton; a second layer that excites localized surface plasmon polariton; a third layer that excites surface phonon polariton; and a fourth layer that excites localized surface plasmon polariton.

6. The apparatus according to any one of the preceding claims, wherein the apparatus has a height between substantially 50 nm to substantially 500 nm.

7. A system for monitoring chemical reactions, the system comprising:an apparatus for monitoring chemical reactions that comprises: at least one layer adapted to excite surface phonon polariton; at least one layer adapted to excite localized surface plasmon polariton; and a substrate; a fluid chamber for receiving chemical reactants; and a radiation source that emits radiation towards the apparatus; wherein the layers are stacked alternately on the substrate, and are further adapted to exhibit a plasmon-phonon coupling mode upon exposure to radiation received from the radiation source to excite the reactants present within the fluid chamber for vibrational fingerprinting of molecules of the reactants undergoing chemical reactions based on their refractive index.

8. The system according to claim 7, further comprising: a radiation detector that receives radiation reflected from the apparatus for converting them into detection data; and a computer interfaced with the radiation detector that operates one or more modules.

9. The system according to claim 8, wherein the computer operates a principal component analysis module, which processes the detection data by performing dimensional reduction thereupon.

10. The system according to claim 9, wherein the computer further operates an identification module that: receives the processed detection data as input; and provides identified molecules within the chemical reactants as output; wherein the identification of molecules is based on classification of vibration signals within their chemical bonds.

11. The system according to any one of claims 7 to 10, wherein the apparatus is disposed within the fluid chamber for its layers to be in substantial contact with the chemical reactants received by the fluid chamber.

12. The system according to any one of claims 7 to 11, wherein the apparatus issubstantially disposed below the radiation source.

13. The system according to any one of claims 7 to 12, wherein the fluid chamber further comprises an inlet and an outlet, for the chemical reactants to be received by the fluid chamber and for the chemical reactants to be directed away from the fluid chamber, respectively.

14. The system according to any one of claims 7 to 13, wherein the layer of the apparatus that is adapted to excite localized surface plasmon polariton is selected from a group of electrically conductive elements that include gold.

15. The system according to any one of claims 7 to 14, wherein the layer of the apparatus that is adapted to excite surface phonon polaritons is selected from a group of polar dielectric materials that include silicon dioxide.

16. The system according to any one of claims 7 to 15, wherein any one of the layers of the apparatus is attached to the substrate, and the substrate comprises material selected from a group of fluoride compounds that include barium fluoride.

17. The system according to any one of claims7 to 16, wherein the apparatus has layers that sequentially comprise: a first layer that excites surface phonon polariton; a second layer that excites localized surface plasmon polariton; a third layer that excites surface phonon polariton; and a fourth layer that excites localized surface plasmon polariton.

18. A method for monitoring chemical reactions, comprising the steps of: preparing an apparatus for monitoring chemical reactions that comprises: at least one layer adapted to excite surface phonon polariton; at least one layer adapted to excite localized surface plasmon polariton; and a substrate; preparing a fluid chamber having chemical reactants therewithin; and emitting radiation towards the apparatus, by a radiation source;wherein the layers are stacked alternately on the substrate, and are further adapted to exhibit a plasmon-phonon coupling mode upon exposure to radiation received from the radiation source to excite the reactants present within the fluid chamber for vibrational fingerprinting of molecules of the reactants undergoing chemical reactions based on their refractive index.

19. The method for monitoring chemical reactions according to claim 18, further comprising the steps of: receiving radiation reflected from the apparatus for conversion into detection data, by a radiation detector; and providing the detection data, by the radiation detector, to a computer that operates one or more modules.

20. The method for monitoring chemical reactions according to claim 19, further comprising the steps of: processing the detection data by performing dimensional reduction thereupon, by a principal component analysis (PCA) module; receiving the processed detection data as input, by an identification module; and providing identified molecules within the chemical reactants as output, by the identification module; wherein the identification of molecules as performed by the identification module is based on classification of vibration signals within their chemical bonds.

Images