Detection method and in-flow detection method

Abstract

A method of multiplexed detection using SES is disclosed. Step 1: providing a SES substrate as a working electrode in an electrochemical cell. Step 2: contacting the SES substrate with a sample comprising a first analyte and a second analyte. Step 3: a cyclic voltammetry measurement step, comprising carrying out a measurement cyclic scan of a potential applied to the SES substrate and performing surface-enhanced spectroscopy on the SES substrate during the measurement cyclic scan. Step 4: a discrimination step, comprising discriminating between the first analyte and the second analyte based on the variation of the SES response of the SES substrate during the measurement cyclic scan. Also disclosed is a method for in-flow SES analysis, wherein the SES substrate comprises a support and a nanoparticle layer of metallic nanoparticles on the support. The nanoparticle layer of the SES substrate is subjected to an oxidative cleaning step, thereby producing an oxide coating at the surfaces of the nanoparticles, and subsequently subjecting the nanoparticle layer to a redefinition step, wherein the oxide coating is removed in the presence of scaffolding ligands so that the scaffolding ligands are arranged between adjacent nanoparticles to define their relative spacing.

Description

[0001] DETECTION METHOD AND IN-FLOW DETECTION METHOD

[0002] The project that lead to this application received funding from EPSRC Ubiquitous Optical Healthcare Technologies (UbOHT) (EP / X037770 / 1 ).

[0003] Field of the Invention

[0004] The present invention relates to methods of multiplex detection using surface-enhanced spectroscopy (SES). The present invention also relates to a method of in-flow SES analysis, including such a method of multiplex detection.

[0005] Background

[0006] Surface-enhanced spectroscopy (SES), including surface enhanced Raman spectroscopy (SERS) and surface-enhanced infrared absorption spectroscopy (SEIRA), has long held promise for rapid, low-cost healthcare sensing, but has yet to fully realise this potential. [1] For example, competitive adsorption of analytes on SERS substrates and the overlapping of characteristic peaks both make multiplexed quantification in complex media, such as bodily fluids, challenging. [2] Electrochemical SES, like electrochemical-SERS (EC-SERS) incorporates the SES substrate into an electrochemical cell, allowing spectra to be recorded under an applied potential. [3, 4]

[0007] EC-SERS has emerged as a promising solution to key challenges in SERS sensing, giving significant improvements in detection limits across a variety of analytes. [3, 5-7] Much EC-SERS work has focused on identifying the optimal applied potential that maximises the SERS signal, and then recording spectra at this potential for analyte detection or quantification. [5-8] However, discussions of the mechanism behind these observed electrochemical enhancements have been vague. These generally focus on increased analyte surface adsorption[9-11], electromagnetic enhancement^, 12], or the removal of organic molecules on the surface

[0013] . Interactions at the surface of nanoparticles, and within the electrochemical double layer (EDL) in particular, are complex and difficult to study or model. [14, 15] In addition to this, previous EC-SERS experiments have often been hampered by variable SERS substrates with ill-defined nanoscale geometries, which make it challenging to study the subtle spectral changes that occur.

[0016]

[0008] Recently, highly reproducible and reusable thin-film SERS substrates have become available from bottom-up self-assembly of AuNPs using a rigid molecular scaffold, cucurbit[n]uril (CB[n], n = 5-8), which yield precise nanogap hotspots. [17, 18] These substrates can be reliably cleaned and regenerated using successive electrochemical oxidation and reduction steps.

[0019]

[0009] The present invention has been devised in light of the above considerations. The development of sensitive, accurate and improved means of multiplex detection using SES is desirable.

[0010] Summary of the Invention

[0011] In the work disclosed here, the reproducible substrates discussed above have allowed the inventors to carefully isolate and identify the effects of applied potential, probing the behaviour of analytes and the EDL within the nanogap. The inventors have found that by making use of a cyclic voltammetry measurement step, including a measurement cyclic scan of applied potential whilst performing surface- enhanced spectroscopy on an SES substrate, a sensitive, accurate and improved means of multiplex detection is provided.

[0012] In a first aspect the present invention provides method of multiplexed detection using SES, comprising, in order: a step of providing a SES substrate as a working electrode in an electrochemical cell; a step of contacting the SES substrate with a sample comprising a first analyte and a second analyte; a cyclic voltammetry measurement step, comprising carrying out a measurement cyclic scan of a potential applied to the SES substrate and performing surface-enhanced spectroscopy on the SES substrate during the measurement cyclic scan; and a discrimination step, comprising discriminating between the first analyte and the second analyte based on the variation of the SES response of the SES substrate during the measurement cyclic scan.

[0013] It may be that the SES is SERS. It may be that the SES is SEIRA.

[0014] The method of multiplexed detection is so-called because such an approach can be capable of the simultaneous detection of one or more components in a multi-component system. In such a system, the presence of a plurality of spectroscopically active analytes makes detection of one or more of the analytes difficult using prior art detection methods.

[0015] The sample comprises (at least) a first analyte and a second analyte. It may be that the sample further comprises a third analyte. It may be that the sample, in total, comprises n analytes, where n is from 2 to 100. It may be that the discrimination step comprises discriminating between a first analyte, and any number of analytes present. For example, when there is also a third analyte, the discrimination step may comprise discriminating between the first analyte, and the second and third analytes.

[0016] The method is carried out using a SES substrate as a working electrode in an electrochemical cell. The SES substrate functions by leveraging plasmonic enhancement within micro- and nano-crevices and depressions on its surface. It may be that the SES substrate is metal based. It may be that the SES substrate comprises a coinage metal. It may be that the SES substrate comprises gold (Au), silver (Ag), aluminium (Al) or copper (Cu).

[0017] It may be that the SES substrate comprises a layer of metallic nanoparticles provided on a support. It may be that the layer of metallic nanoparticles comprises one or more monolayers of nanoparticles, and such a SES substrate is herein described as a monolayer aggregate (MLagg; an aggregate of monolayers). It may be that the metallic nanoparticles are selected from one or more metals which support optical- or mid-infrared-frequency surface plasmons. It may be that the metallic nanoparticles are formed of gold, silver, aluminium or copper. It may be that the metallic nanoparticles have an average diameter that is greater than or equal to 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, or 75 nm. It may be that the metallic nanoparticles have an average diameter that is less than or equal to 5000, 4000, 3000, 2000, 1000, 500, 250, 150, or 100 nm. The foregoing lower and upper limits may be freely combined to generate a range. It may be that the metallic nanoparticles have an average diameter that is from 15 to 5000 nm, 25 to 1000 nm, 50 to 250 nm, or about 80nm. It may be that the support comprises glass, fluorine tin oxide-coated glass, gold and / or silicon. It may be that the nanoparticle layer comprises a region with 1 , 2, 3, 4, 5 or 6 monolayers of metallic nanoparticles. It may be that the nanoparticle layer comprises 1 , 2, 3, 4, 5 or 6 monolayers of metallic nanoparticles, or some combination of regions each independently comprising 1 , 2, 3, 4, 5 or 6 monolayers of metallic nanoparticles. It may be that the metallic nanoparticles are stabilised by scaffolding ligands in the gaps between adjacent nanoparticles. It may be that the scaffolding ligands are one or more molecules selected from the group consisting of cucurbit[n]uril, a polystyrene molecule, a thiol molecule, 3-mercaptopropionic acid, citrate, acetic acid, cysteamine, dopamine, paracetamol, ethanol and methanol. It may be that the nanoparticle layer is provided by any suitable route based on self-assembly of nanoparticles. It may be that the nanoparticles are substantially spherical.

[0018] The SES substrate is the working electrode in an electrochemical cell. As such, the SES substrate is electrically conductive at least to the extent needed so that a potential can be applied to the SES active surface of the SES substrate. As will be understood from the disclosure above the electrochemical cell has optical access for SES spectroscopy.

[0019] It may be that the electrochemical cell comprises a counter electrode and a reference electrode (i.e. , a three-electrode cell). It may be that the electrochemical cell comprises a counter electrode (i.e., a two- electrode cell). It may be that the electrochemical cell is a potentiostat. The counter electrode is conductive and inert. It may be that the counter electrode is a metallic wire. It may be that the counter electrode comprises platinum (Pt), copper (Cu), gold (Au), or silver (Ag). It may be that the counter electrode comprises FTO-coated glass, carbon (graphite), stainless steel, or glassy carbon. It may be that the reference electrode comprises silver. It may be that the reference electrode is a Ag / AgCI electrode. It may be that the reference electrode is a standard hydrogen electrode, a saturated calomel electrode, a copper-copper(ll) sulphate electrode or a mercury / mercuric oxide electrode.

[0020] The method comprises contacting the SES substrate with a sample comprising a first analyte and a second analyte. It may be that the sample is applied in discrete amounts, batch-style. It may be that the sample is applied continuously, in-flow style, where the sample flows over and past the SES substrate. The electrochemical cell, of which the SES substrate is part, comprises electrolyte which contacts the SES substrate, the reference electrode, and the counter electrode. It may be that the electrolyte is provided as part of the sample, or seperately to the same.

[0021] It may be that the electrolyte is present at a concentration that is greater than or equal to 0.1 , 0.5, 1 , 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 100, or 250 mM. It may be that the electrolyte is present at a concentration that is less than or equal to 10000, 1000, 500, 250, 150, 100, or 75 mM. The foregoing lower and upper bounds may be combined freely, where physically reasonable, to generate a range. The electrolyte may be present at a concentration in the range of from 0.1 mM to 1 M, from 1 mM to 100 mM, from 25 mM to 75 mM, or about 50 mM. It may be that the electrolyte is present at a concentration in the range of from 0.1 mM to 10 M, from 1 mM to 1 M, from 250 mM to 750 mM, or about 500 mM. The strength of the SES response is increased with increasing electrolyte concentration. Without wishing to be bound by theory, it is though that this is due to decreasing Debye length and EDL thickness, making it easier for analytes to infiltrate the EDL and bind to the SES substrate surface. It may be that the electrolyte comprises potassium phosphate, sodium phosphate, lithium acetate, sodium sulphate, magnesium sulphate, sodium perchlorate, sodium nitrate, sodium chloride, potassium nitrate, sodium fluoride, or magnesium perchlorate in aqueous solution. It may be that a non-aqueous ionic solvent is used.

[0022] The method comprises a cyclic voltammetry measurement step, in which a measurement cyclic scan of potential applied to the SES substrate is carried out, whilst performing surface-enhanced spectroscopy on the SES substrate.

[0023] As used herein, the “measurement cyclic scan” refers to a scan which comprises less than one, one, or more than one cycles of applied potential variation. As such, it may be that the measurement cyclic scan comprises a single cycle of potential variation. It may be that the measurement cyclic scan comprises two or more cycles of potential variation. It may be that the measurement cyclic scan comprises 2, 3, 4, 5, 6, 7, 8, 9 or 10 cycles of potential variation. It may be that a cycle comprises the application of a variable potential that moves or is altered from a starting value, between an upper (more positive) bound, and a lower (more negative) bound. It may be that the variable applied potential finishes a cycle at or near the same potential as the starting value. In other embodiments, the variable applied potential finishes a cycle at a different potential to the starting value.

[0024] The applied potential may be varied in a continuous and / or a discrete way. When varied in a continuous way, it may be that the scan rate is constant throughout a cycle, or it may vary. It may be that the measurement cyclic scan comprises periods of continuous change in applied potential, and discrete steps in applied potential. It may be that the upper bound is a positive potential. It may be that the positive potential of the upper bound is from 0.1 to 1.5V, or from 0.25 to 1 V, or about 0.5V. It may be that the lower bound is a negative potential. It may be that the negative potential of the lower bound is from -0.1 to -2V, or from -0.5 to -1.5V, or about -1 V. It may be that the upper bound is a positive potential and the lower bound is a negative potential. It may be that the measurement cyclic scan has an average scan rate of from 5 to 1000 mV / s, or from 25 to 250 mV / s, or about 50 mV / s.

[0025] Within a measurement cyclic scan, when there are 2 or more cycles, it may be that adjacent cycles follow the same potential pattern, or a different potential pattern. It may be that the upper and lower bounds vary between cycles. It may be that the scan rate varies between cycles.

[0026] The method comprises a discrimination step, comprising discriminating between the first analyte and the second analyte based on the variation of the SES response of the SES substrate during the measurement cyclic scan. It may be that discrimination comprises concluding that both the first analyte and the second analyte are present. It may be that the discrimination step comprises identifying the first and / or the second analyte. It may be that the discrimination step comprises identifying the first analyte. It may be that the discrimination step comprises identifying the second analyte.

[0027] The discrimination step can be carried out by any suitable means capable of discriminating between the first and second analytes based on the variation of the SES response of the SES substrate. It may be that the discrimination step is carried out manually (or augmented), through comparison between the variation of the SES response of the SES substrate and prior measured responses of known analytes. It may be that the discrimination step is carried out, or augmented, in an automated fashion, using a computer program. It may be that the discrimination step is carried out, or augmented, using a machine learning algorithm.

[0028] As used herein, reference to the variation of a feature in the SES response, or the SES response perse (i.e. a portion or the whole of the spectrum), during the measurement cyclic scan, may refer to the variation between two or more discrete instances of the feature or the response, at one or more discrete applied potentials, or it may refer to the continuous change observed in the feature or the response throughout the entire measurement cyclic scan.

[0029] When discussing the variation of a feature in the SES response, or the SES response perse (i.e. a portion or the whole of the spectrum), during the measurement cyclic scan, it may be that the variation of the feature or the response is considered over only one cycle, or it may be considered over multiple cycles, and in the latter case suitable normalization methods may be used to compare different cycles. As such, it may be that a feature or a response being considered is a normalized feature or response. It may be that normalization is relative to an element of the spectrum which produces a substantially constant response throughout the measurement cyclic scan, irrespective of the number of cycles, for example a fixed element of the substrate, such as a scaffolding ligand in the case of the MLaggs described below. Normalization may also be used when comparing signals produced by different SES substrates, or by different measurement cyclic scans.

[0030] It may be that the discrimination step is based on the variation in a compound feature of the SES response, generated by combining two or more of the characteristic features of a peak in the SES response described herein. For example, a difference between a maximum and a minimum in the peak area of a peak during the measurement cyclic scan; or the difference between the applied potential at which the peak area is at a maximum, and the applied potential at which the peak position is at a maximum, during the measurement cyclic scan. A suitable machine learning algorithm can be used to analyse numerous and more complex features of a portion of the SES response (e.g. a peak, or a number of peaks within a certain window), or the entire SES response.

[0031] Reference herein to the position of a SES response feature (i.e. a peak) is understood to be a reference to the spectral position of that feature (i.e. the spectral position of the peak).

[0032] In some embodiments the discrimination step further comprises discriminating between the first analyte and the second analyte based on the variation of a peak in the SES response during the measurement cyclic scan.

[0033] It may be that the peak is produced by the first analyte. It may be that the peak is produced by the second analyte. Advantageously, peaks in the SES response each vary and oscillate in characteristic ways during the measurement cyclic scan, allowing the variation in a peak to be used in the discrimination step.

[0034] In some embodiments the discrimination step further comprises discriminating between the first analyte and the second analyte based on the variation in the peak area of a peak in the SES response during the measurement cyclic scan. It may be that the peak is produced by the first analyte. It may be that the peak is produced by the second analyte. The peak area (the “integral”) of a peak may be directly related to the strength of the response of an active vibrational mode in an analyte contacted on the SES substrate. It is found that the peak areas for peaks in the SES response each vary and oscillate in characteristic ways during the measurement cyclic scan, allowing the variation in peak area to be used in the discrimination step.

[0035] The peak area may be calculated using suitable software. It may be that peak height is used, instead of or in addition to peak area, and peak height may be measured by suitable software, or by visual inspection of spectra.

[0036] In some embodiments the discrimination step further comprises discriminating between the first analyte and the second analyte based on the applied potential at which the peak area of a peak in the SES response is at a maximum during the measurement cyclic scan.

[0037] It may be that the peak is produced by the first analyte. It may be that the peak is produced by the second analyte.

[0038] The peak area for a peak in the SES response may be found to vary, and to pass through a maximum value at a characteristic value of applied potential, unique to the active mode of a given analyte. It is therefore possible that the applied potential at which the peak area is at a maximum can be used in the discrimination step. It may be that the applied potential at which the peak area of a peak in the SES response is at a minimum is used in the discrimination step.

[0039] It may be that the discrimination step comprises discrimination between the first and the second analyte based on the shape of a peak in the SES response. It may be that the discrimination step comprises discrimination between the first and the second analyte based on the full width at half maximum (FWHM) value of a peak in the SES response. It may be that the FWHM value is used in place of or in addition to the peak area in the foregoing discussion; for example, the applied potential at which a FWHM value of a peak is at a maximum during the measurement cyclic scan can be used in the discrimination step.

[0040] FWHM may be measured by suitable software, or by visual inspection of spectra.

[0041] In some embodiments the discrimination step further comprises discriminating between the first analyte and the second analyte based on the variation of the position of a peak in the SES response during the measurement cyclic scan.

[0042] It may be that the peak is produced by the first analyte. It may be that the peak is produced by the second analyte.

[0043] The position of a peak in the SES response is related to the energy level structure, for example vibrational energy level structure in the case of SERS and SEIRA, of the analyte producing the peak. It is found that the position of each peak in the SES response varies and oscillates in characteristic ways during the measurement cyclic scan, allowing the variation in the peak position to be used in the discrimination step.

[0044] Peak position may be measured by suitable software, or by visual inspection of spectra. In some embodiments the discrimination step further comprises discriminating between the first analyte and the second analyte based on the applied potential at which the position of a peak in the SES response is at a maximum value, and / or the applied potential at which the position of a peak is at a minimum value, during the measurement cyclic scan.

[0045] It may be that the peak is produced by the first analyte. It may be that the peak is produced by the second analyte.

[0046] The position for a peak in the SES response may be found to vary, and to pass through a maximum and minimum values at characteristic values of applied potentials, unique to the active mode of a given analyte. It is therefore possible that the applied potential at which the position of a peak is at a maximum and / or the applied potential at which the position of a peak is at a minimum can be used in the discrimination step.

[0047] In some embodiments the discrimination step further comprises discriminating between the first analyte and the second analyte based on the difference between the SES response when scanning in a positive direction and when scanning in a negative direction during the measurement cyclic scan.

[0048] It may be that the SES response exhibits a difference in SERS response (i.e. hysteresis) between scans in the positive direction (towards a more positive applied potential) and scans in the negative direction (towards a more negative applied potential). That is, the SES response varies in a non-symmetric way when scanning in a positive direction and when scanning in a negative direction.

[0049] It may be found that the presence and extent of hysteresis in the SES response varies in a unique way for a given analyte, or mixture of analytes. As such, the presence and extent of hysteresis in the SES response can be used in the discrimination step.

[0050] In some embodiments the discrimination step further comprises discriminating between the first analyte and the second analyte based on the applied potential at which the peak area of a peak in the SES response is at a maximum when scanning in each of

[0051] (i) a positive direction, and

[0052] (ii) a negative direction, during the measurement cyclic scan.

[0053] It may be that the peak is produced by the first analyte. It may be that the peak is produced by the second analyte.

[0054] It is found that the applied potential at which the area of a peak in the SES response is at a maximum can be different when scanning in a positive direction, and when scanning in a negative direction during the measurement cyclic scan. As such, the presence and extent of the difference in this applied potential can be used in the discrimination step.

[0055] It may be that peak height is used instead of or in addition to the peak area, and so the applied potential at which the height of a peak in the SES response is at a maximum when scanning in each of the positive direction and the negative direction is used in the discrimination step. It may be that FWHM is used instead of or in addition to the peak area, and so the applied potential at which the FWHM of a peak in the SES response is at a maximum when scanning in each of the positive direction and the negative direction is used in the discrimination step.

[0056] In some embodiments the discrimination step further comprises discriminating between the first analyte and the second analyte based on the applied potential at which the position of a peak in the SES response is at a maximum, and / or the applied potential at which the position of the peak is at a minimum, when scanning in each of

[0057] (i) a positive direction, and

[0058] (ii) a negative direction, during the measurement cyclic scan.

[0059] It may be that the peak is produced by the first analyte. It may be that the peak is produced by the second analyte.

[0060] It may be found that the applied potential at which the position of a peak in the SES response is at a maximum can be different when scanning in a positive direction, and when scanning in a negative direction, during the measurement cyclic scan. As such, the presence and extent of the difference in this applied potential can be used in the discrimination step.

[0061] In some embodiments the discrimination step further comprises discriminating between the first analyte and the second analyte based on the variation between a first peak and a second peak in the SES response during the measurement cyclic scan.

[0062] It may be found that individual peaks in the SES response vary uniquely relative to one another during the measurement cyclic scan, and so discrimination between a first analyte and a second analyte can be carried out based on the difference in behaviour between a first peak and a second peak during the measurement cyclic scan. It may be that both of the first peak and the second peak are produced by the first or second analyte, or it may be that the first peak is produced by the first analyte, and the second peak is produced by the second analyte.

[0063] It may be that the discrimination step comprises discriminating between the first analyte and the second analyte based on the variation between a first peak and a second peak in any of the foregoing peak features (peak area, peak shape, peak height, FWHM, peak position).

[0064] It may be that the discrimination step comprises discriminating between the first analyte and the second analyte based on the variation between the area, the height or the FWHM of the first peak and the second peak during the measurement cyclic scan.

[0065] It may be that the discrimination step comprises discriminating between the first analyte and the second analyte based on the variation between the applied potentials at which the area, the height or the FWHM of the first peak and the second peak are at a maximum, or where they are at a minimum.

[0066] It may be that the discrimination step comprises discriminating between the first analyte and the second analyte based on the variation between the position of the first peak and the second peak. It may be that the discrimination step comprises discriminating between the first analyte and the second analyte based on the variation between the applied potentials at which the area, the height or the FWHM of the first peak and the second peak are at a maximum, or where they are at a minimum.

[0067] In some embodiments the method further comprises a conditioning step prior to the cyclic voltammetry measurement step, the conditioning step comprising a conditioning cyclic scan of potential applied to the SES substrate between a positive potential and a negative potential in the presence of the sample.

[0068] It may be found that by performing a conditioning cyclic scan on the SES substrate, the response of the first and / or second analyte during the cyclic voltammetry measurement step increases in strength.

[0069] Without wishing to be bound by theory, it is thought that the conditioning cyclic scan repeatedly inverts the EDL which has the effect of drawing the analyte molecules closer to the surface of the SES substrate, leading to greater SES enhancement. Each cycle of the conditioning cyclic scan starts at a starting potential, cycles between a positive and a negative potential, and finishes back at or near the starting potential. In some embodiments, no spectroscopy is performed during the conditioning cyclic scan.

[0070] It may be that the conditioning cyclic scan comprises one cycle. It may be that the conditioning cyclic scan comprises two or more cycles. It may be that the conditioning cyclic scan comprises 2, 3, 4, 5, 6, 7, 8, 9 or 10 cycles. It may be that the positive potential is from 0.1 to 1.5V, or from 0.25 to 1V, or about 0.5V. It may be that the negative potential is from -0.1 to -2V, or from -0.5 to -1.5V, or about -1 V. It may be that the conditioning cyclic scan has an average scan rate of from 5 to 500 mV / s, or from 25 to 250 mV / s, or about 50 mV / s. It may be that the conditioning cyclic scan has an average scan rate > 50 mV / s, or > 500 mV / s.

[0071] It may be that the intensity of a peak in the SES response increases by > 100%, > 200%, > 300%, or > 400% after 5 cycles of the conditioning cyclic scan.

[0072] In some embodiments the method further comprises a holding step, prior to the conditioning step, in the presence of the sample, wherein a static potential is applied for a period of time.

[0073] Performing a holding step increases the SES enhancement seen following a conditioning step. It may be that the holding step is performed at open circuit potential (OCP). It may be that the holding step comprises applying the static potential for a period of from 1 minute to 60 minutes, or from 5 minutes to 30 minutes, or about 10 minutes.

[0074] In some embodiments the measurement cyclic scan comprises varying the applied potential continuously.

[0075] Continuous variation may be pseudo-continuous, where a substantially continuous change is achieved through rapid and small steps in potential; such is the case when a continuous change is implemented digitally. It may be that the measurement cyclic scan comprises varying the applied potential discretely. For example, it may be that the applied potential is varied discretely in steps of a specified size, where each applied potential is applied for some not insignificant amount of time, for example in steps that are from 0.01 to 1 .5 V in size, or from 0.1 to 1 V in size, or from 0.2 to 0.5V or about 0.25V in size.

[0076] In some embodiments the measurement cyclic scan is carried out at an average scan rate of > 5 mV / s. Performing the measurement cyclic scan at scan rates > 5 mV / s increases the level of hysteresis observed in the SES response, which can aid discrimination.

[0077] It may be that the measurement cyclic scan is carried out at an average scan rate of > 10 mV / s, > 50 mV / s, > 100 mV / s, > 250 mV / s or > 500 mV / s.

[0078] In some embodiments the sample comprises a sensitising electrolyte capable of forming a complex with the first and / or second analyte.

[0079] Discrimination between analytes can be improved by the inclusion of a sensitising electrolyte. Interaction of analyte and sensitising electrolyte generates additional SES features, enabling improved differentiation between analytes. In the case of SERS, this is especially true for analytes with inherently weak Raman cross-sections, such as non-aromatic molecules.

[0080] It may be that the sensitising electrolyte comprises a metal ion. It may be that the metal ion is an iron (Fe) ion, or a copper (Cu) ion. It may be that the ion is Fe(lll) or Cu(ll). It may be that the sensitising electrolyte comprises sulphate ions. It may be that the sensitising electrolyte is present at from 1 mM to 500 mM, or 5 mM to 100 mM or about 25 mM, or about 10 mM or about 50 mM. It may be that the sensitising electrolyte produces an ionic cation or anion layer on top of the SES substrate surface. It may be that the sensitising electrolyte forms a complex with an analyte in bulk solution, and then adsorbs onto the SES substrate (i.e. , into a hotspot).

[0081] In a second aspect the present invention provides a method for in-flow SES analysis, comprising: performing the method of multiplexed detection of the first aspect, wherein the SES substrate comprises a support and a nanoparticle layer of metallic nanoparticles on the support; then subjecting the nanoparticle layer of the SES substrate to an oxidative cleaning step, thereby producing an oxide coating at the surfaces of the nanoparticles, and subsequently subjecting the nanoparticle layer to a redefinition step, wherein the oxide coating is removed in the presence of scaffolding ligands so that the scaffolding ligands are arranged between adjacent nanoparticles to define their relative spacing.

[0082] Herein, the second step comprising oxidative cleaning and redefinition is known as “ReSERS”. The method for in-flow sensing allows for numerous complex mixtures of analytes to be processed using the method of multiplex detection herein in series. The presence of the ReSERS step cleans and recycles the SES substrate, ready for use again. The method of in-flow sensing may comprise a further (or more than one further) round of multiplexed detection and cleaning and redefinition.

[0083] The invention includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.

[0084] Summary of the Figures

[0085] Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures in which: Fig. 1 illustrates cyclic electrochemical SERS of adenine using an MLagg substrate. 80 nm AuNPs selfassembled with cucurbit[5]uril (CB[5]) and deposited onto FTO-coated glass substrate. SERS spectro- electrochemical cell consists of MLagg-CB[5] substrate (working electrode), Pt wire (counter electrode) and Ag / AgCI (reference electrode).

[0086] Fig. 2 Time-series SERS spectra (1 s integration time, 785 nm 1 mW laser) of the MLagg cycled between +0.5 V and -1 V in 10 pM adenine (ADN) and 50 mM potassium phosphate buffer (pH 7.0) at 50 mV s-1for 5 cycles (top). Normalised peak area for ADN peak at -732 cm'1(middle), applied potential (bottom, dashed), and current (bottom, solid) plotted vs cycle (bottom).

[0087] Fig. 3 SERS spectra of MLagg after initial in-situ electrochemical cleaning and regeneration (bottom line), after addition of 10 pM ADN (at open-circuit potential, middle line), and after 5 EC cycles (taken at -0.7V, top line). ADN peak at -732 cm'1indicated by *, inset shows neutral ADN (dominates at pH 7.0).

[0088] Fig. 4 Time-series SERS spectra (zoomed-in to narrow windows of Raman shift) of (top) 10 pM adenine (ADN) and (bottom) 10 pM cytosine (CYT), conditions and cycling as Fig. 2.

[0089] Fig. 5 Normalised SERS peak area vs applied potential (vs. Ag / AgCI) for ring-breathing modes of (top) ADN at -732 cm'1and (bottom) CYT at -796 cm'1when cycled as in Fig. 4. Intensity of marker colour denotes the time elapsed (darker = longer), arrows indicate cycle direction. First cycle is removed for clarity.

[0090] Fig. 6 Peak frequency vs applied potential for same peaks as Fig. 5. Intensity of marker colour denotes the time elapsed (darker = longer), arrows indicate cycle direction. First cycle is removed for clarity.

[0091] Fig. 7 Time-series SERS spectra of 10 pM ADN, conditions and cycling as Fig. 2. Arrows mark three additional ADN peaks at -970 cm'1(bottom), -1320 cm'1(middle), -1370 cm1(top).

[0092] Fig. 8 Final cycle signals vs potential (vs. Ag / AgCI) for the additional ADN peaks marked with arrows in Fig. 7 (positive scans = 0, negative scans = o).

[0093] Fig. 9 ADN peak area (-732 cm'1, normalised by CB[5]) plotted against EC cycles for different potassium phosphate buffer (pH 7.0) concentrations. Lines from bottom to top: 5 mM; 50 mM; 100 mM; 500 mM.

[0094] Fig. 10 Schematic potential-induced molecular reorientation of adenine (ADN) on Au NP surface.

[0095] Fig. 11 Schematic of analyte molecules infiltrating the EDL during potential cycling.

[0096] Fig. 12 Molecular structures of neutral adenine (ADN), cytosine (CYT), guanine (GUA) and thymine (THY).

[0097] Fig. 13 Concentration series for detection of (top) ADN and (bottom) CYT. Peak areas (ADN -732 cm'1, CYT -796 cm'1, normalised by CB[5]) are extracted before (open markers) and after (filled markers) 5 EC cycles (as Fig. 2) in 500 mM potassium phosphate buffer (pH 7.0). Error bars from standard deviation of 3 measurements. Langmuir-Hill model fit, with limit of detection calculated from intersection with 3a confidence band of noise level (grey). Fig. 14 Time-series SERS spectra while cycling MLagg in 5 pM ADN, 50 pM CYT, 50 pM GUA and 50 pM THY in 500 mM KPB (pH 7.0). GUA, ADN, and CYT peaks at 660, 732, 796 cm1respectively. THY peak at -1340 cm'1in Figs. 32-35.

[0098] Fig. 15 Final cycle signals from Fig. 14 vs potential (vs. Ag / AgCI) for each nucleobase.

[0099] Fig. 16 Scanning electron micrographs (SEM) of the MLagg-CB[5] after initial cleaning and regeneration.

[0100] Fig. 17 Schematic diagram of the electrochemical SERS (EC-SERS) set-up including a cross-section diagram of the spectro-electrochemical cell. CE = counter electrode, RE = reference electrode, and WE = working electrode.

[0101] Fig. 18 Cyclic voltammetry in different electrolytes. Overlaid cyclic voltammograms of Mlaggs in (top) 50 mM potassium phosphate buffer (pH 7.0) and (bottom) 10 pM adenine (ADN) and 50 mM potassium phosphate buffer (pH 7.0). The applied potential was cycled from 0 V (vs. Ag / AgCI) to +0.5 V, down to -1 V and then back to 0 V five times at a scan rate of 50 mV s-1 .

[0102] Fig. 19 Cyclic electrochemical SERS of potassium phosphate buffer using an MLagg substrate. Timeseries SERS spectra (1 s integration time, 785 nm 1 mW laser) of the MLagg cycled between +0.5 V and -1 V in 50 mM potassium phosphate buffer (pH 7.0) at 50 mV s-1for 5 cycles.

[0103] Fig. 20 EC-SERS response of CB[5]. (a, c) Peak frequency vs applied potential (vs. Ag / AgCI) for the peaks at (a) -828 cm'1and (c) -456 cm'1, which are attributed to the CB[5] in the scaffolded MLagg, conditions and cycling as Fig. 19. (b,d) Normalised SERS peak area vs applied potential for the same peaks as (a,c). Intensity of marker colour denotes the time elapsed (darker = longer), arrows indicate cycle direction.

[0104] Fig. 21 Oxidation of gold during EC cycling, (a) Time-series SERS spectra (zoomed-in) of a 50 mM potassium phosphate buffer solution (pH 7.0), conditions and cycling as Fig. 19. (b) Normalised SERS peak area vs applied potential (vs. Ag / AgCI) for the gold oxide peak at -555 cm'1. Intensity of marker colour denotes the time elapsed (darker = longer), arrows indicate cycle direction.

[0105] Fig. 22 Cyclic electrochemical SERS of adenine at different scan rates. Time-series SERS spectra (1 s integration time, 785 nm 1 mW laser) of the MLagg cycled between +0.5 V and -1 V in 10 pM adenine (ADN) and 50 mM potassium phosphate buffer (pH 7.0) at 500 mV s-1for 5 cycles. Normalised peak area for ADN peak at -732 cm'1(middle), applied potential (bottom, dashed), and current (bottom, solid) plotted vs time.

[0106] Fig. 23 Cyclic electrochemical SERS of adenine at different scan rates. Time-series SERS spectra (1 s integration time, 785 nm 1 mW laser) of the MLagg cycled between +0.5 V and -1 V in 10 pM adenine (ADN) and 50 mM potassium phosphate buffer (pH 7.0) at 5 mV s-1 for 5 cycles. Normalised peak area for ADN peak at -732 cm'1(middle), applied potential (bottom, dashed), and current (bottom, solid) plotted vs time.

[0107] Fig. 24 Comparison of ADN peak signals during 5 mV / s scan. Normalised SERS peak area vs applied potential (vs. Ag / AgCI) for the peaks at (a) -732 cm'1(ring-breathing), (b) -1370 cm'1(N-C-H in-plane bending), (c) -1320 cm'1(C-N stretch), (d) -970 cm'1(5-ring deformation). Conditions and cycling as Fig. 23, intensity of marker colour denotes the time elapsed (darker = longer). Arrows denote cycle direction.

[0108] Fig. 25 Comparison of ADN peak positions during 5 mV / s scan. SERS peak center vs applied potential (vs. Ag / AgCI) for the peaks at (a) -732 cm'1(ring-breathing), (b) -1370 cm'1(N-C-H in-plane bending), (c) -1320 cm'1(C-N stretch), (d) -970 cm'1(5-ring deformation). Conditions and cycling as Fig. 23, intensity of marker colour denotes the time elapsed (darker = longer). Arrows denote cycling direction.

[0109] Fig. 26 Comparison of different EC conditions. ADN peak area (-732 cm'1, normalised by CB[5]) plotted against time for different EC conditions: waiting at open circuit potential (OCP), applying a constant potential of -0.7V, cycling between +0.5 V and -1 V at 50 mV / s and 500 mV / s. The electrolyte was 10 pM ADN and 50 mM potassium phosphate buffer (pH 7.0).

[0110] Fig. 27 Cyclic electrochemical SERS of cytosine. Time-series SERS spectra (1 s integration time, 785 nm 1 mW laser) of the MLagg cycled between +0.5 V and -1 V in 10 pM cytosine (CYT) and 50 mM potassium phosphate buffer (pH 7.0) at 50 mV s-1for 5 cycles. Normalised peak area for CYT peak at -796 cm'1(middle), applied potential (bottom, dashed), and current (bottom, solid) plotted vs time.

[0111] Fig. 28 EC-SERS response of cytosine. (a,c,e) Normalised SERS peak area vs applied potential (vs. Ag / AgCI) for the peaks at (a) -796 cm'1(ring-breathing), (c) -1040 cm'1(ring deformation), and (e) -1310 cm'1(ring stretch C-N).(b,d,f) Peak frequency vs applied potential for same peaks as (a,c,e). Intensity of marker colour denotes the time elapsed (darker = longer), conditions and cycling as Fig. 27.

[0112] Fig. 29 Effect of buffer concentration on cyclic EC-SERS response. ADN peak area (-732 cm'1, normalised by CB[5]) plotted against time for different potassium phosphate buffer (pH 7.0) concentrations. Conditions and cycling as Fig. 2.

[0113] Fig. 30 Calculated Debye lengths for different buffer concentrations. Debye lengths calculated using Debye-Huckel Theory

[0032] for different potassium phosphate buffer (pH 7.0) concentrations.

[0114] Fig. 31 Effect of waiting at OCP before performing cyclic EC-SERS. ADN peak area (-732 cm'1, normalised by CB[5]) for different times spent waiting at OCP (0.2V vs. Ag / AgCI) before cycling the potential. Conditions and cycling as Fig. 2.

[0115] Fig. 32 Cyclic electrochemical SERS of guanine. Time-series SERS spectra (1 s integration time, 785 nm 1 mW laser) of the MLagg cycled between +0.5 V and -1 V in 100 pM guanine (GUA) and 50 mM potassium phosphate buffer (pH 7.0) at 50 mV s-1for 5 cycles. Normalised peak area for GUA peak at -660 cm'1(middle), applied potential (bottom, dashed), and current (bottom, solid) plotted vs time.

[0116] Fig. 33 EC-SERS response of guanine, (a) Normalised SERS peak area vs applied potential (vs. Ag / AgCI) for the peak at -660 cm'1(ring-breathing), (b) Peak frequency vs applied potential for the same peaks as (a). Intensity of marker colour denotes the time elapsed (darker = longer), conditions and cycling as Fig. 32. Arrows denote cycle direction.

[0117] Fig. 34 Cyclic electrochemical SERS of thymine. Time-series SERS spectra (1 s integration time, 785 nm

[0118] 1 mW laser) of the MLagg cycled between +0.5 V and -1 V in 100 pM thymine (THY) and 50 mM potassium phosphate buffer (pH 7.0) at 50 mV s-1for 5 cycles. Normalised peak area for THY peak at -1340 cm'1(middle), applied potential (bottom, dashed), and current (bottom, solid) plotted vs time.

[0119] Fig. 35 EC-SERS response of thymine, (a, c) Normalised SERS peak area vs applied potential (vs. Ag / AgCI) for the peaks at (a) -785 cm'1(ring-breathing) and (c) -1340 cm'1(CH3 bend), (b, d) Peak frequency vs applied potential for the same peaks as (a). Intensity of marker colour denotes the time elapsed (darker = longer), conditions and cycling as Fig. 34. Arrows denote cycle direction.

[0120] Fig. 36 Cyclic electrochemical SERS of multiple DNA nucleobases. Time-series SERS spectra (1 s integration time, 785 nm 1 mW laser) of the MLagg cycled between +0.5 V and -1 V in 50 pM ADN, 50 pM CYT and 50 pM GUA and 500 mM potassium phosphate buffer (pH 7.0) at 50 mV s-1for 5 cycles. Applied potential (bottom, dashed), and current (bottom, solid) plotted vs time.

[0121] Fig. 37 Effect of buffer concentration on cyclic EC-SERS response, (top left) ADN and (top right) CYT peak area (normalised by CB[5]) plotted against time for different potassium phosphate buffer (pH 7.0) concentrations, (bottom left) ADN and (bottom right) CYT maximum peak area (normalised by CB[5]) plotted against cycle number for different potassium phosphate buffer (pH 7.0) concentrations.

[0122] Fig. 38 Maximum peak areas after 5 cycles for (left) ADN and (right) CYT (normalised by CB[5]) plotted against KPB concentration.

[0123] Fig. 39 Comparison of the signal hysteresis at different scan rates. Normalised SERS peak area vs applied potential (vs. Ag / AgCI) for ADN and CYT recorded at 50 mV / s and 5 mV / s. 5 EC cycles were performed for each test and the KPB concentration was 50 mM. Intensity of marker colour denotes the time elapsed (darker = longer). Arrows indicate cycle direction.

[0124] Fig. 40 Comparison of the peak position hysteresis at different scan rates. Normalised SERS peak position vs applied potential (vs. Ag / AgCI) for ADN and CYT recorded at 50 mV / s and 5 mV / s. 5 EC cycles were performed for each test and the KPB concentration was 50 mM. Intensity of marker colour denotes the time elapsed (darker = longer). Arrows indicate cycle direction.

[0125] Fig. 41 Comparison of different scan rates. ADN peak area (-732 cm'1, normalised by CB[5]) plotted against time for different cyclic voltammetry scan rates: 5 mV / s, 50 mV / s and 500 mV / s. Conditions and cycling as Fig. 26.

[0126] Fig. 42 Sensitisation of pyruvate detection with metal ion complexation. EC-SERS spectra of electrolyte only in different electrolytes and applied potentials (vs. Ag / AgCI): (a) 50 mM Na2SC>4 pH 7.0, (b) 50 mM CuSC pH 7.0, and (c) 10 mM Fe2(SC>4)3 adjusted to pH 3.0 with H2SO4 (low pH minimises hydrolysis of Fe(lll)).

[0127] Fig. 43 Sensitisation of pyruvate detection with metal ion complexation. EC-SERS spectra of 1 mM pyruvate in different electrolytes and applied potentials (vs. Ag / AgCI): (a) 50 mM Na2SC>4 pH 7.0, (b) 50 mM CuSC pH 7.0, and (c) 10 mM Fe2(SCU)3 adjusted to pH 3.0 with H2SO4 (low pH minimises hydrolysis of Fe(lll)). Peaks associated with pyruvate are highlighted. Dotted line in (c) distinguished pyruvate- associated peak at 1020 cm'1from peak related to SCU2' (995 cm'1). Fig. 44 Sensitisation of branch-chain amino acid detection with Cu(ll). EC-SERS spectra of 1 mM L- leucine in different electrolytes and applied potentials (vs. Ag / AgCI): (a) 50 mM Na2SC>4 pH 7.0, (b) 50 mM CuSC pH 7.0. Peaks associated with the amino acids are highlighted.

[0128] Fig. 45 Sensitisation of branch-chain amino acid detection with Cu(ll). EC-SERS spectra of 1 mM L- valine in different electrolytes and applied potentials (vs. Ag / AgCI): (a) 50 mM Na2SC>4 pH 7.0, (b) 50 mM CuSC pH 7.0. Peaks associated with the amino acids are highlighted.

[0129] Fig. 46 Structure of hypoxanthine, HYP, (top). Time-series SERS spectra (1 s integration time, 785 nm 1 mW laser) of the MLagg cycled between +0.5 V and -1 V in 100 pM HYP and 500 mM potassium phosphate buffer (pH 7.0) at 50 mV s-1for 5 cycles (top). Normalised peak area for HYP peak at -730 cm-1(middle), applied potential (bottom, dashed), and current (bottom, solid) plotted vs cycle (bottom).

[0130] Fig. 47 Time-series SERS spectra (zoomed-in to narrow windows of Raman shift) of 100 pM hypoxanthine (HYP), conditions and cycling as Fig. 46.

[0131] Fig. 48 Normalised SERS peak area vs applied potential (vs. Ag / AgCI) for (left) HYP peak at -730 cm'1when cycled as in Fig. 46. Peak frequency vs applied potential (right) for same peak as left. Intensity of marker colour denotes the time elapsed (darker = longer), arrows indicate cycle direction.

[0132] Detailed Description of the Invention

[0133] Aspects and embodiments of the present invention will now be discussed with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference. The below discussion focusses on SERS substrates, and SERS measurements, however the skilled person understands that the below discussion also applies to other SES substrates and measurements, such as SEIRA.

[0134] Here, the inventors study the EC-SERS behaviour of DNA nucleobases. Despite being well-studied[20, 21], there is still debate over the nature of their Au surface adsorption[22-24] and EC-SERS behaviour

[0025] . The choice of a set of four similar but subtly different molecules allows the inventors to examine the EC-SERS enhancement mechanism in detail. It is shown that by cycling the electrochemical potential one can significantly increase the SERS signal of nucleobases, dramatically reducing their limits of detection. For each analyte, unique oscillations in, for example, the intensity and position of their characteristic vibrational peaks are observed, which, it is thought, result from the potential-induced molecular reorientation of the molecules and their interaction with the EDL. These distinct EC fingerprints, produced by the unique oscillations, enable the simultaneous multiplexed detection of the four nucleobases, opening up new possibilities for label-free biomarker detection in complex fluids.

[0135] To fabricate reliable EC-SERS substrates with precise sub-1 nm nanogap hotspots, the inventors selfassemble citrate-stabilized 80 nm AuNPs using CB[5], a barrel-shaped molecule which binds AuNPs together through its carbonyl portals.

[0026] The AuNP:CB[5] assembly is concentrated and deposited as a thin film onto conductive FTO-coated glass.

[0017] These close-packed 1-2 monolayer AuNP aggregates (termed monolayer aggregates or ‘MLaggs’) are used as the working electrode in a spectro- electrochemical cell (Fig. 1 ).

[0136] It is possible to reliably clean and regenerate the MLaggs using a recently developed in-situ electrochemical SERS hotspot regeneration scheme (‘ReSERS’)-

[0019] First, the applied potential is stepped to +1.5V to strip the ligand-stabilised nanogaps of molecular adsorbates and form a meta-stable gold oxide plug. Then, the hotspots are regenerated by stepping the potential to -0.8V in a solution containing the CB[5] scaffolding molecule, which reduces the oxide layer and restabilises the nanogaps. The cleaning procedure can be repeated many (>30) times without damaging the substrate. This creates reliable EC-SERS substrates (<5% relative standard deviation) with large enhancement factors (>106) enabling the inventors to carefully study the electrochemical behaviour of analytes in the nanogap.

[0019]

[0137] Adenine (ADN) is chosen as a model analyte as it has been extensively studied with SERS[20, 21], but questions remain over its adsorption on Au[22-24] and its EC-SERS behaviour

[0025] . SERS spectra are recorded by illuminating the MLagg with a 785 nm laser through the transparent FTO-coated glass. SERS spectra of the clean MLaggs recorded at open-circuit potential (OCP - 0.2V vs. Ag / AgCI) in 50 mM potassium phosphate buffer (KPB) (pH 7.0) show the two characteristic CB[5] peaks at 456, 828 cm'1(Fig. 3).

[0019] When 10 pM ADN is added, a small peak is observed at 732 cm-1, assigned to the symmetric ring breathing mode coupled with its in-plane NH2 bend (Fig. 3).

[0021] At pH 7.0, ADN is expected to be primarily neutral, and can be deprotonated at its pKa of 9.8.

[0138] The inventors perform cyclic voltammetry in this 10 pM ADN, 50 mM KPB solution whilst recording SERS spectra of the MLagg. The potential is swept from 0V to +0.5 V, reversed to -1 V, and then back to 0V at 50 mV s-1for 5 cycles. Clear changes are observed in the EC-SERS spectra of the MLagg (Fig. 2), with small shifts in the CB[5] peak positions (<2 cm'1) and intensity increase by -30% at positive potentials (Fig. 20). Peaks at 350, 555 cm'1associated with the oxidation of gold

[0019] also start to appear above +0.25V (Fig. 21 ). Much more significant changes are observed for the ADN peaks, particularly at 732 cm'1(VADN), where the intensity increases more than 400% in 5 cycles (Fig. 2).

[0139] Consistent oscillations are seen within each EC cycle in the magnitude and position of VADN (Fig. 4, top). As the potential is scanned anodically from -1V, the intensity of ADN increases (Fig. 5, top) and the centre of the peak starts to shift to higher wavenumbers (Fig. 6, top). The intensity of VADN reaches a maximum around -0.1V and then plateaus but the centre of the peak continues to shift, reaching 739 cm'1at +0.5V. During the cathodic scan, the magnitude starts to increase again and a maximum is observed at lnaY=- 0.6V which is -25% larger than during the positive scan. As the negative scan proceeds to -1 V, the peak intensity dramatically reduces and the peak centre returns to 732 cm'1. The positive and negative scans show a clear hysteresis in both the intensity and position of VADN . Interestingly, this hysteresis is noticeably reduced when the scan rate is reduced to 5 mV / s (Figs. 24 and 25). These observations are key to understanding the EC cycling enhancement.

[0140] The fourfold increase in VADN observed over 5 EC cycles cannot be achieved by simply applying a constant potential for 5 minutes (or longer) (Fig. 26). Many EC-SERS studies have focused on finding the optimal potential, and then using spectra recorded at this potential for detection / quantification.[5-7] However, it is found that cycling the potential for 5 minutes results in peaks that are >200% of those recorded after applying a constant potential (-0.7V) for the same time (Fig. 26). If the scan rate is increased to 500 mV / s, it speeds up this process, achieving the same signal >10x faster than applying a constant potential (Fig. 26).

[0141] Performing the same cycling with cytosine (CYT), another nucleobase, reveals similar general behaviour (Fig. 4, bottom). Its characteristic peak at 796 cm'1(VCYT, ring breathing mode

[0021] ) also increases over 5 cycles with hysteretic oscillations in its magnitude and position within each cycle (Fig. 5, bottom and Fig. 6, bottom). However consistent differences are observed, with the signal maxima at V(nax=-0.6V for ADN and -0.4V for CYT (Fig. 5). This means cycling the potential can selectively enhance specific peaks for particular analytes of interest, opening up significant possibilities for multiplexed detection in mixed fluids.

[0142] Besides the peak shifts in primary ADN and CYT peaks, oscillations are also observed in other characteristic peaks (Fig. 7 and Fig. 8). Interestingly, these peaks do not all change in the same way, with each vibrational mode exhibiting different shifts in peak position and intensity as the potential is cycled. For ADN, the signal maxima for -970 cm'1(5-ring deformation), -1318 cm'1(C-N stretch), and -1370 cm'1(N-C-H in-plane bending) modes are observed at V(nax=+0.5V, -0.9V and +0.25V respectively (Fig. 8). As is discussed below, this does not match charge-transfer or deprotonation mechanisms (pKa is uncorrelated with V^), and instead suggests EDL-driven reorientation. It is also noted that the vibrational Stark effect, where a molecular vibrational energy is modulated by an external electric field, can cause shifts in both the magnitude and position of SERS peaks. [27, 28] Previous studies of this effect with EC- SERS reported peak shifts -3 cm'1for nitrile-containing self-assembled monolayers (SAMs) and intensity increases <50%.

[0028] This is much smaller than the changes observed here for nucleobases (peak shifts >15 cm'1).

[0143] The inventors thus tentatively attribute the signal oscillations to potential-induced molecular reorientation of analytes at the Au surface (Fig. 10). As Raman scattering is proportional to the fourth power of the polarizability component in the direction of the optical field (perpendicular to the Au surface), potential- induced molecular reorientation will result in large SERS intensity changes, providing the polarizability tensor of the vibrational mode is anisotropic.

[0029] Electrochemical Tip-Enhanced Raman Spectroscopy (EC-TERS) indeed shows that the ADN 732 cm'1(ring-breathing) mode is supressed when the molecules lie flat on a Au surface and enhanced when the ADN is vertically orientated.

[0025] The inventors thus propose that at -1V, the ADN molecules lie flat on the surface, starting to tilt as the potential increases, becoming vertical at -0.6V, and continuing to tilt while adsorbed via the amino group. During negative scans, this process is reversed (Fig. 7 and Fig. 8). Even at low analyte densities, the same behaviour is observed, suggesting there is no interaction between analyte molecules.

[0144] This potential-induced molecular reorientation model is supported by the minimal spectral shifts in the CB[5] peaks (Fig. 20). CB[5] strongly binds to Au via its carbonyl-rimmed portals, meaning it cannot tilt on the surface in the same way as the nucleobases.

[0030] The small peak shifts (< 2 cm-1) observed are thus due to residual vibrational Stark effects from the EDL which immerses them. Evidence for further effects from the EDL come from the consistent hysteresis observed in analyte signal peak and frequency (Fig. 4, Fig. 5, and Fig. 6). Since analyte polarisation is induced by the internal electric field of the EDL in which they are immersed, this can cause the observed symmetry breaking of reorientation with an energy barrier for tilting the molecules. When the scan rate is reduced, there is more time to overcome this energy barrier, resulting in the reduced hysteresis observed (Figs. 24 and 25). Molecular reorientation can thus account for the oscillations seen within each EC cycle. However, a sustained increase in signal is seen over multiple cycles, which cannot simply be explained by diffusion (Fig. 26). The inventors thus propose that this is due to analyte molecules infiltrating the EDL and binding onto the AuNP surface. At OCP, the EDL is dominated by absorbed anions from the KPB electrolyte, as the surface is positively charged. The layer of adsorbed anions and water molecules prevents analyte molecules from binding to the surface, which sit above the EDL. As the potential goes negative, the EDL internal field reverses and the analyte molecules on its top surface are drawn inside the EDL, polarised, and twist to eventually bind onto the Au irreversibly. EC-cycling thus grabs analyte molecules, twists them through the EDL, and collects them under the EDL on the Au, accumulating them continuously (Fig. 11 ). When the potential is kept constant, this mechanism does not operate, restricting the signal.

[0145] To investigate EDL infiltration further, EC-SERS cyclic voltammetry of ADN and CYT is performed in KPB buffer (pH 7.0) of different concentration (Fig. 9, Fig. 29). It is found that the analyte signal consistently increases with buffer concentration, with 300% signal increase between 5mM to 500mM KPB after 5 cycles (Fig. 9). Simultaneously, the Debye length and EDL thickness decrease (Fig. 30), making it easier for analytes to infiltrate, explaining why more analyte molecules reach the Au surface. As expected from this model, since the CB[5] molecules are bound onto the Au surface, no significant increase in CB[5] signal is expected or seen (<30% over 5 cycles). Summarising these observations, the inventors propose that potential cycling enables analyte molecules to travel through the EDL and reach the Au surface, resulting in the sustained increase in SERS signals. Potential-induced molecular reorientation is responsible for the oscillations observed in intensity and frequency of analyte peaks.

[0146] For SERS to be effective in low-cost rapid healthcare sensing, it must be able to quantify very low concentrations of biomolecules and detect multiple analytes simultaneously. Cyclic EC-SERS can improve both these capabilities. The inventors first investigate the limits of detection (LoDs) of nucleobases using this system. Cyclic voltammetry of solutions containing 0.05-100 pM ADN or CYT in 500 mM KPB is performed whilst recording SERS spectra of the MLagg. Before introducing analyte, the MLagg is cleaned and scaffolded using the ReSERS procedure.

[0019] The analyte solution is added to the EC-SERS cell and then left at OCP for 10 minutes (Fig. 31). The potential is then cycled between +0.5V and -1 V at 50 mV s-1for 5 cycles. 3 EC-SERS measurements are recorded for each solution. To compare measurements on different MLaggs, the spectra are normalised using the CB[5] 828 cm'1peak from spectra recorded post-ReSERS cleaning but before adding analyte.

[0147] Cyclic EC-SERS consistently increases the magnitude of the characteristic nucleobase peaks and enables the detection of concentrations which have no visible analyte peaks at OCP. To determine the limits of detection, the CB[5] normalised peak areas (ADN-732 cm'1, CYT-796 cm'1) are plotted against time and the maximum after 5 cycles extracted. The peak area at for each concentration is then fit to a Langmuir-Hill model (see Methodology below), and the LoD determined from its intersection with the 3a confidence band of the noise level. The LoD for ADN after cyclic EC-SERS is 147 nM compared to 3.7 pM before cycling the potential. For CYT, the LoD is 260 nM with cyclic EC-SERS and 1.7 pM without (Fig. 13).

[0148] Finally, multiplexed detection with all four DNA nucleobases ADN, CYT, guanine (GUA) and thymine (THY) is performed (Fig. 12). Analyte solutions containing 50 pM of CYT, GUA, THY and 5 pM ADN in 500 mM KPB are cycled between +0.5V and -1V at 50 mV s-1for 5 cycles. A lower ADN concentration is used as its large Raman cross section dominates EC-SERS spectra for equal analyte concentrations (Fig. 36).

[0031] The inventors observe and distinguish between the peaks of all the nucleobases (Fig. 14). As for ADN and CYT, GUA and THY exhibit their own unique peak oscillations (Fig. 15). Interestingly, the two purine nucleobases (ADN and GUA) display similar behaviour (1iax--0.4V) as do the two pyrimidine nucleobases (CYT and THY) with ,1:IX-0V. The inventors suggest this is because of the way that the molecules of different length and polarizability are twisted through and penetrate the EDL, as it does not correspond to pKa or redox states. Indeed, a much larger class of analytes can now be screened through this assay, and their spectro-electrochemical maps analysed by more advanced data analysis including machine learning, to develop both a deeper understanding of ionic double-layer interactions with analytes, as well as improved distinction and discrimination between analytes. Embodiments of the present invention have also been used to successfully characterise uracil, dopamine, epinephrine, norepinephrine, serotonin and tryptophan through their unique potential dependent SERS responses.

[0149] In conclusion, it has been demonstrated that cyclic EC-SERS can significantly reduce the limits of detection of nucleobases (down to <0.15 pM so far) and enable their simultaneous multiplexed detection. Using low-cost, reliable, reusable SERS substrates, the inventors have studied the unique EC-SERS behaviours of each of the tested molecules. The inventors propose a mechanism for the EC-SERS enhancement based on the potential-induced reorientation of the molecules and their penetration through the EDL barrier. This approach offers a simple way to overcome some of the challenges that have plagued SERS sensing and opens up new possibilities for multiplexed detection in mixed fluids.

[0150] Enhancing discrimination

[0151] Electrolyte concentration

[0152] The inventors performed EC-SERS cyclic voltammetry of adenine (ADN) and cytosine (CYT) in potassium phosphate buffer (KPB) (pH 7.0) of different concentrations (Fig. 37 and 38). It is found that the analyte signal consistently increases with buffer concentration, with 300% signal increase between 5mM to 500mM KPB after 5 cycles. It is believed that this is because the Debye length and electrochemical double layer (EDL) thickness decrease with increasing electrolyte concentration (Fig. 30), making it easier for analytes to infiltrate the EDL and bind to the Au surface. This results in larger SERS signal.

[0153] Scan rate

[0154] The inventors have studied the cyclic EC-SERS behaviour of ADN and CYT at three different scan rates (5 mV / s, 50 mV / s and 500 mV / s). The same general behaviours are observed at each scan rate: there is an overall increase in the SERS signal with EC cycling, and there are potential dependent oscillations observed in the intensity and frequency of analyte peaks. However, there are some notable differences which can be used to optimise the signal, and enhance discrimination.

[0155] We see a clear hysteresis between the positive and negative scans in relation to the intensity and position of the analyte peaks (Figs. 39 and 40). This hysteresis is due to the configuration / orientation of the analyte in the EDL, which does not exactly follow the instantaneous potential and instead depends on the history. When the scan rate is reduced to 5 mV / s from 50 mV / s, the hysteresis is reduced. The inventors propose this is because there is an energy barrier to the reorientation of the molecules and when the scan rate is reduced, there is more time to overcome this barrier.

[0156] The inventors have also investigated how we can use the scan rate to increase the speed at which analytes are trapped. At a scan rate of 50 mV / s, it is found that over 5 cycles (5 minutes), the ADN peak increased by more than 400%. If the scan rate is increased to 500 mV / s, it speeds up this process (Figs. 26 and 41 ). This approach can achieve the same signal >10x faster than applying a constant potential (- 0.7V) and reduce the cycling time by a least half.

[0157] Metal ion sensitisation

[0158] The discrimination of analytes via cyclic voltammetry and SERS is enhanced by the inclusion of metal ions such as Fe(lll) and Cu(ll), which interact with analytes to form complexes. These interactions generate additional EC-SERS features, enabling improved differentiation between analytes. This is especially relevant for analytes with inherently weak Raman cross-sections, such as non-aromatic molecules.

[0159] For instance, when Fe(lll) or Cu(ll) ions are present, an analyte such as pyruvate can exhibit distinct changes in both their electrochemical behaviour and SERS spectra (Figs. 42 and 43). These changes arise from metal ion coordination, which shifts redox potentials and introduces new vibrational modes associated with metal-analyte complexes. Notably, structurally similar molecules such as L-leucine and L- valine can exhibit different spectral features when complexing with Cu(ll) ions (Figs. 44 and 45). Thus, sensitisation with metal ions not only enhances the selectivity and sensitivity of analyte detection but also provides deeper insights into ion-molecule interactions at the electrode surface.

[0160] Multiplex detection with cleaning

[0161] A protocol for carrying out in-flow multiplex detection repeatedly, on changing samples, is:

[0162] 1. Cleaning and regenerating the as-fabricated SERS substrate (ReSERS)

[0163] 2. Performing multiplexed cyclic EC-SERS measurements as described herein

[0164] 3. Cleaning and regenerating the SERS substrate (ReSERS) to remove previous analytes

[0165] 4. Performing multiplexed cyclic EC-SERS measurements as described herein

[0166] To clean and regenerate the SERS substrate (MLagg), 50 mM potassium phosphate buffer (pH 7.0) is injected (either via pipetting or continuously flowing solution in a microfluidic set-up) into the EC-SERS cell and a potential of +1.5 V vs Ag / AgCI is applied for 60 s. After cleaning, the original buffer solution is removed and 1 mM CB[5] in 50 mM potassium phosphate buffer (pH 7.0) is injected to the cell. A potential of -0.80 V vs Ag / AgCI is then applied for 15 s. The EC-SERS cell is then washed out with buffer before performing the multiplexed cyclic EC-SERS measurement. Successful cleaning is confirmed by measuring SERS spectra.

[0167] To perform the multiplexed cyclic EC-SERS measurement, the test solution (consisting of multiple analytes and 50 mM potassium phosphate buffer (pH 7.0)) is injected into the EC-SERS cell. The electrochemical potential is then cycled between +0.5 V and -1 V vs Ag / AgCI at 50 mV s-1for 5 cycles (other parameters can also be used). SERS spectra are recorded simultaneously. The EC-SERS cell is then washed out with buffer before performing the ReSERS cleaning again and repeating the process.

[0168] Discrimination between similar molecules

[0169] The present invention can be used to discriminate between analyte molecules with similar structures, and therefore similar static SERS responses. For example, hypoxanthine (Fig. 46) and adenine (Fig. 2) have similar SERS peaks that show maximum responses at similar potentials, but can be differentiated between using the potential-induced shifts in their peak positions (ADN - Fig. 4 (top), Fig. 5 (top), Fig. 6 (top); HYP - Fig. 47, Fig. 48).

[0170] Methodology

[0171] Materials

[0172] All chemicals were used as received. Citrate-stabilized 80 nm AuNPs (optical density 1.0 at 555 nm) were purchased from BBI Solutions. Analytical-grade chloroform (> 99.8 %) was obtained from Merck.

[0173] K2HPO4 (^98%), and KH2PO4 (^98%) were from Alfa Aesar. Cucurbit[5]uril hydrate (= 20% water), adenine (ADN, >99%), cytosine (CYT, >99%), guanine (GUA, >99%) and thymine (THY, >99%) were obtained from Sigma-Aldrich. Polydimethylsiloxane (PDMS) was prepared using a SYLGARD 184 kit from DOWSIL (Dow Silicones). Fluorine-doped tin oxide (FTO)-coated glass slides (TEC 10) were purchased from Ossila Ltd and were cleaned and cut to 10 x 15 mm2slides prior to use. All aqueous solutions were prepared using deionized (DI) water (>18.2-MQ cm-1) from a Purelab Ultra Scientific water purification system.

[0174] Monolayer aggregate preparation

[0175] MLagg substrates were prepared by mixing 500 pL citrate-stabilised 80 nm AuNPs with equal volume chloroform, and initiating aggregation with the addition of 20 pL of 2 mM CB[5].[17, 18] Aggregation was facilitated with 1 min of vigorous shaking, after which the aggregates settled at the liquid-liquid interface. Excess ligands and salts were removed by replacing the aqueous supernatant with fresh DI water. This washing step was repeated three times. The aggregates were then concentrated by carefully decreasing the volume of the aqueous phase to ~20 pL. The AuNP aggregate was then transferred via pipette to a cleaned FTO-coated glass slide and allowed to air dry.

[0176] SERS

[0177] SERS measurements were recorded on a custom-built Raman set-up with an Andor Newton 970 EMCCD camera coupled to a Shamrock 168 spectrometer and a Matchbox 785 nm diode laser (Fig. 17). Excitation and collection were performed through an Olympus LUMPIanFI / IR *40 W NA 0.80 waterimmersion objective (in inverted configuration) at 1 s integration times with 1 mW laser power.

[0019]

[0178] EC-SERS

[0179] An EC-SERS cell was fabricated from PDMS to accommodate a three-electrode electrochemical system: a Pt wire (Sigma-Aldrich) counter electrode, a leakless Ag / AgCl / KCI (LF-1-45 from Innovative Instruments Ltd.) reference electrode, and a MLagg SERS substrate on FTO-coated glass as the working electrode (Figs. 16 and 17).

[0019] The electrolyte compartment was defined using a 6 mm diameter biopsy punch. Using custom 3D-printed stage holders, the EC-SERS cell was sealed and mounted onto the stage of an inverted Raman set-up, with SERS probed from below the cell. Electrochemical measurements were conducted using a portable potentiostat (Rodeostat) from IO Rodeo. All potentials were referenced to the Ag / AgCI reference electrode.

[0180] Cleaning and regeneration of Mlaggs

[0181] To clean and regenerate the MLagg

[0019] , 50 mM potassium phosphate buffer (pH 7.0) was pipetted into the EC-SERS cell and a potential of +1.5 V vs Ag / AgCI was applied for 60 s. After cleaning, the original buffer solution was removed and 1 mM CB[5] in 50 mM potassium phosphate buffer (pH 7.0) was pipetted to the cell. A potential of -0.80 V vs Ag / AgCI was then applied for 15 s. The EC-SERS cell was then washing out with buffer before performing the next measurement. If traces of previously detected analyte were evident from the SERS spectrum, another round of cleaning / regeneration was conducted.

[0182] SEM measurements

[0183] Scanning electron microscope (SEM) imaging of MLaggs deposited on FTO-coated glass was conducted using a FEI Philips Dualbeam Quanta 3D SEM (dwell 3-10 s, HV 2 kV, current 50 pA, and =2.0 mm WD).

[0184] Data analysis

[0185] SERS spectra were background-corrected using asymmetric least squares (ALS) baseline correction. Analyte peak areas and center points were determined by defining a spectral region and then fitting Gaussian curves to the peaks of interest. CB[5] normalised peak areas were calculated using the 828 cm-1CB[5] peak from the clean rescaffolded MLagg spectra recorded before adding the analyte of interest.

[0186] Langmuir-Hill model

[0187] The concentration series in Fig. 13 were fitting according to the standard Langmuir-Hill equation: where Kais the analyte concentration to occupy half of the binding sites (dissociation constant), n is the Hill coefficient, |C| is the analyte concentration, and A and b are constants. These were fitted to Fig. 13 and the fit coefficients as well as the LoD are detailed in Table S1 .

[0188] Table S1: Langmuir-Hill fit coefficients for adenine (Fig. 13, top) and cytosine (Fig. 13, bottom). The LoD is calculated as 3a above the noise level.

[0189] ***

[0190] The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for obtaining the disclosed results, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof.

[0191] While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention.

[0192] For the avoidance of any doubt, any theoretical explanations provided herein are provided for the purposes of improving the understanding of a reader. The inventors do not wish to be bound by any of these theoretical explanations.

[0193] Any section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described.

[0194] Throughout this specification, including the claims which follow, unless the context requires otherwise, the word “comprise” and “include”, and variations such as “comprises”, “comprising”, and “including” will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

[0195] It must be noted that, as used in the specification and the appended claims, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from “about” one particular value, and / or to “about” another particular value. When such a range is expressed, another embodiment includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent “about,” it will be understood that the particular value forms another embodiment. The term “about” in relation to a numerical value is optional and means for example + / - 10%. References

[0196] A number of publications are cited above in order to more fully describe and disclose the invention and the state of the art to which the invention pertains. Full citations for these references are provided below. The entirety of each of these references is incorporated herein.

[0197] [1] Langer, J.; Jimenez de Aberasturi, D.; Aizpurua, J.; Alvarez-Puebla, R. A.; Auguie, B.; Baumberg, J. J.; Bazan, G. C.; Bell, S. E. J.; Boisen, A.; Brolo, A. G.; Choo, J.; Cialla-May, D.; Deckert, V.; Fabris, L.; Faulds, K.; Garcia de Abajo, F. J.; Goodacre, R.; Graham, D.; Haes, A. J.; Haynes, C. L; Huck, C.; Itoh, T.; Kall, M.; Kneipp, J.; Kotov, N. A.; Kuang, H.; Le Ru, E. C.; Lee, H. K.; Li, J.- F.; Ling, X. Y.; Maier, S. A.; Mayerhbfer, T.; Moskovits, M.; Murakoshi, K.; Nam, J.-M.; Nie, S.; Ozaki, Y.; Pastoriza-Santos, I.; Perez-Juste, J.; Popp, J.; Pucci, A.; Reich, S.; Ren, B.; Schatz, G. C.; Shegai, T.; Schlucker, S.; Tay, L.-L.; Thomas, K. G.; Tian, Z.-Q.; Van Duyne, R. P.; Vo-Dinh, T.; Wang, Y.; Willets, K. A.; Xu, C.; Xu, H.; Xu, Y.; Yamamoto, Y. S.; Zhao, B.; Liz-Marzan, L. M. Present and Future of Surface-Enhanced Raman Scattering. ACS Nano 2020, 14 (1), 28-117. https: / / doi.org / 10.1021 / acsnano.9b04224.

[0198] [2] Sloan-Dennison, S.; Wallace, G. Q.; Hassanain, W. A.; Laing, S.; Faulds, K.; Graham, D. Advancing SERS as a Quantitative Technique: Challenges, Considerations, and Correlative Approaches to Aid Validation. Nano Converg. 2024, 11 (1), 33. https: / / doi.org / 10.1186 / s40580- 024-00443-4.

[0199] [3] Brosseau, C. L.; Colina, A.; Perales-Rondon, J. V.; Wilson, A. J.; Joshi, P. B.; Ren, B.; Wang, X. Electrochemical Surface-Enhanced Raman Spectroscopy. Nat. Rev. Methods Primer 2023, 3 (1), 1-21. https: / / doi.org / 10.1038 / s43586-023-00263-6.

[0200] [4] Wu, D.-Y.; Li, J.-F.; Ren, B.; Tian, Z.-Q. Electrochemical Surface-Enhanced Raman Spectroscopy of Nanostructures. Chem. Soc. Rev. 2008, 37 (5), 1025-1041. https: / / doi.org / 10.1039 / B707872M.

[0201] [5] Zaleski, S.; Clark, K. A.; Smith, M. M.; Eilert, J. Y.; Doty, M.; Van Duyne, R. P. Identification and Quantification of Intravenous Therapy Drugs Using Normal Raman Spectroscopy and Electrochemical Surface-Enhanced Raman Spectroscopy. Anal. Chem. 2017, 89 (4), 2497-2504. https: / / doi.org / 10.1021 / acs.analchem.6b04636.

[0202] [6] Zhao, L.; Blackburn, J.; Brosseau, C. L. Quantitative Detection of Uric Acid by Electrochemical- Surface Enhanced Raman Spectroscopy Using a Multilayered Au / Ag Substrate. Anal. Chem. 2015, 87 (1), 441—447. https: / / doi.org / 10.1021 / ac503967s.

[0203] [7] Greene, B. H. C.; Alhatab, D. S.; Pye, C. C.; Brosseau, C. L. Electrochemical-Surface Enhanced Raman Spectroscopic (EC-SERS) Study of 6-Thiouric Acid: A Metabolite of the Chemotherapy Drug Azathioprine. J. Phys. Chem. C 2017, 121 (14), 8084-8090. https: / / doi.Org / 10.1021 / acs.jpcc.7b01179.

[0204] [8] Chotoye, S. A. B.; Eisnor, M. M.; Ball, R. B. E.; Brosseau, C. L. Development of an Electrochemical Surface-Enhanced Raman Spectroscopic Biosensor for the Direct Detection of Glutathione. J. Raman Spectrosc. 2023, 54 (6), 587-595. https: / / doi.org / 10.1002 / jrs.6522.

[0205] [9] Markin, A. V.; Arzhanukhina, A. I.; Markina, N. E.; Goryacheva, I. Y. Analytical Performance of Electrochemical Surface-Enhanced Raman Spectroscopy: A Critical Review. TrAC Trends Anal. Chem. 2022, 157, 116776. https: / / doi.Org / 10.1016 / j.trac.2022.116776.

[0206]

[0010] Willets, K. A. Probing Nanoscale Interfaces with Electrochemical Surface-Enhanced Raman Scattering. Curr. Opin. Electrochem. 2019, 13, 18-24. https: / / d0i.0rg / l 0.1016 / j.coelec.2018.10.005.

[0207]

[0011] Oyamada, N.; Minamimoto, H.; Murakoshi, K. In Situ Observation of Unique Bianalyte Molecular Behaviors at the Gap of a Single Metal Nanodimer Structure via Electrochemical Surface- Enhanced Raman Scattering Measurements. J. Phys. Chem. C 2019, 123 (40), 24740-24745. https: / / doi.org / 10.1021 / acs.jpcc.9b07361 .

[0208]

[0012] Aranda, D.; Garcia-Gonzalez, F.; Avila Ferrer, F. J.; Lopez-Tocon, I.; Soto, J.; Otero, J. C. Computational Model for Electrochemical Surface-Enhanced Raman Scattering: Key Role of the Surface Charges and Synergy between Electromagnetic and Charge-Transfer Enhancement Mechanisms. J. Chem. Theory Comput. 2022, 78 (11), 6802-6815. https: / / doi.org / 10.1021 / acs.jctc.2c00633.

[0209]

[0013] Lynk, T. P.; Sit, C. S.; Brosseau, C. L. Electrochemical Surface-Enhanced Raman Spectroscopy as a Platform for Bacterial Detection and Identification. Anal. Chem. 2018, 90 (21), 12639-12646. https: / / doi.org / 10.1021 / acs.analchem.8b02806.

[0210]

[0014] Shin, S.-J.; Kim, D. FL; Bae, G.; Ringe, S.; Choi, FL; Lim, H.-K.; Choi, C. FL; Kim, H. On the Importance of the Electric Double Layer Structure in Aqueous Electrocatalysis. Nat. Commun. 2022, 73 (1), 174. https: / / doi.org / 10.1038 / s41467-021-27909-x.

[0211]

[0015] Schmickler, W. Double Layer Theory. J. Solid State Electrochem. 2020, 24 (9), 2175-2176. https: / / d0i.0rg / l 0.1007 / s10008-020-04597-z.

[0212]

[0016] Jones, T. Fabrication of Nanostructured Electrodes for Electrochemical Surface-Enhanced Raman Spectroscopy (E-SERS): A Review. Mater. Sci. Technol. 2023, 39 (16), 2287-2301. https: / / doi.org / 10.1080 / 02670836.2023.2197729.

[0213]

[0017] Grys, D.-B.; Niihori, M.; Arul, R.; Sibug-Torres, S. M.; Wyatt, E. W.; de Nijs, B.; Baumberg, J. J. Controlling Atomic-Scale Restructuring and Cleaning of Gold Nanogap Multilayers for Surface- Enhanced Raman Scattering Sensing. ACS Sens. 2023, 8 (7), 2879-2888. https: / / doi.org / 10.1021 / acssensors.3c00967.

[0018] Niihori, M.; Fbldes, T.; Readman, C. A.; Arul, R.; Grys, D.-B.; Nijs, B. de; Rosta, E.; Baumberg, J. J. SERS Sensing of Dopamine with Fe(lll)-Sensitized Nanogaps in Recleanable AuNP Monolayer Films. Small 2023, 19 (48), 2302531 . https: / / doi.org / 10.1002 / smll.202302531 .

[0214]

[0019] Sibug-Torres, S. M.; Grys, D.-B.; Kang, G.; Niihori, M.; Wyatt, E.; Spiesshofer, N.; Ruane, A.; de Nijs, B.; Baumberg, J. J. In Situ Electrochemical Regeneration of Nanogap Hotspots for Continuously Reusable Ultrathin SERS Sensors. Nat. Commun. 2024, 15 (1), 2022. https: / / doi.Org / 10.1038 / S41467-024-46097-y.

[0215]

[0020] Davison, G.; Jones, T.; Liu, J.; Kim, J.; Yin, Y.; Kim, D.; Chio, W.-l. K.; Parkin, I. P.; Jeong, H.-H.; Lee, T.-C. Computer-Aided Design and Analysis of Spectrally Aligned Hybrid Plasmonic Nanojunctions for SERS Detection of Nucleobases. Adv. Mater. Technol. 2023, 8 (7), 2201400. https: / / d0i.0rg / l 0.1002 / admt.202201400.

[0216]

[0021] Madzharova, F.; Heiner, Z.; Giihlke, M.; Kneipp, J. Surface-Enhanced Hyper-Raman Spectra of Adenine, Guanine, Cytosine, Thymine, and Uracil. J. Phys. Chem. C 2016, 120 (28), 15415- 15423. https: / / doi.org / 10.1021 / acs.jpcc.6b02753.

[0217]

[0022] Harroun, S. G. The Controversial Orientation of Adenine on Gold and Silver. ChemPhysChem 2018, 79 (9), 1003-1015. https: / / doi.org / 10.1002 / cphc.201701223.

[0218]

[0023] Kundu, J.; Neumann, O.; Janesko, B. G.; Zhang, D.; Lal, S.; Barhoumi, A.; Scuseria, G. E.;

[0219] Halas, N. J. Adenine- and Adenosine Monophosphate (AMP)-Gold Binding Interactions Studied by Surface-Enhanced Raman and Infrared Spectroscopies. J. Phys. Chem. C 2009, 773 (32), 14390-14397. https: / / doi.org / 10.1021 / jp903126f.

[0220]

[0024] Yoshimoto, T.; Seki, M.; Okabe, H.; Matsuda, N.; Wu, D.; Futamata, M. Three Distinct Adsorbed States of Adenine on Gold Nanoparticles Depending on pH in Aqueous Solutions. Chem. Phys. Lett. 2022, 786, 139202. https: / / doi.Org / 10.1016 / j.cplett.2021.139202.

[0221]

[0025] Martin Sabanes, N.; Ohto, T.; Andrienko, D.; Nagata, Y.; Domke, K. F. Electrochemical TERS Elucidates Potential-Induced Molecular Reorientation of Adenine / Au(111). Angew. Chem. 2017, 729 (33), 9928-9933. https: / / doi.org / 10.1002 / ange.201704460.

[0222]

[0026] Taylor, R. W.; Lee, T.-C.; Scherman, O. A.; Esteban, R.; Aizpurua, J.; Huang, F. M.; Baumberg, J. J.; Mahajan, S. Precise Subnanometer Plasmonic Junctions for SERS within Gold Nanoparticle Assemblies Using Cucurbit[ n ] Uril “Glue.” ACS Nano 2011 , 5 (5), 3878-3887. https: / / doi.org / 10.1021 / nn200250v.

[0223]

[0027] Bishop, D. M. The Vibrational Stark Effect. J. Chem. Phys. 1993, 98 (4), 3179-3184. https: / / d0i.0rg / l 0.1063 / 1.464090.

[0224]

[0028] Wright, D.; Sangtarash, S.; Mueller, N. S.; Lin, Q.; Sadeghi, H.; Baumberg, J. J. Vibrational Stark Effects: Ionic Influence on Local Fields. J. Phys. Chem. Lett. 2022, 73 (22), 4905-4911. https: / / doi.org / 10.1021 / acs.jpclett.2c01048.

[0029] Schlucker, S. Surface-Enhanced Raman Spectroscopy: Concepts and Chemical Applications.

[0225] Angew. Chem. Int. Ed. 2014, 53 (19), 4756-4795. https: / / doi.org / 10.1002 / anie.201205748.

[0226]

[0030] Grys, D.-B.; de Nijs, B.; Salmon, A. R.; Huang, J.; Wang, W.; Chen, W.-H.; Scherman, O. A.; Baumberg, J. J. Citrate Coordination and Bridging of Gold Nanoparticles: The Role of Gold Adatoms in AuNP Aging. ACS Nano 2020, 14 (7), 8689-8696. https: / / doi.org / 10.1021 / acsnano.0c03050.

[0227]

[0031] Barhoumi, A.; Zhang, D.; Tam, F.; Halas, N. J. Surface-Enhanced Raman Spectroscopy of DNA.

[0228] J. Am. Chem. Soc. 2008, 130 (16), 5523-5529. https: / / doi.org / 10.1021 / ja800023j.

[0229]

[0032] Aqueous Electrolyte Solutions. Fundamentals of Electrochemistry; John Wiley & Sons, Ltd, 2005; pp 99-126. https: / / doi.org / 10.1002 / 047174199X.ch7.

Claims

28Claims:1 . A method of multiplexed detection using SES, comprising, in order: a step of providing a SES substrate as a working electrode in an electrochemical cell; a step of contacting the SES substrate with a sample comprising a first analyte and a second analyte; a cyclic voltammetry measurement step, comprising carrying out a measurement cyclic scan of a potential applied to the SES substrate and performing surface-enhanced spectroscopy on the SES substrate during the measurement cyclic scan; and a discrimination step, comprising discriminating between the first analyte and the second analyte based on the variation of the SES response of the SES substrate during the measurement cyclic scan.

2. The method of claim 1 , wherein the discrimination step further comprises discriminating between the first analyte and the second analyte based on the variation of a peak in the SES response during the measurement cyclic scan.

3. The method of claim 1 or claim 2, wherein the discrimination step further comprises discriminating between the first analyte and the second analyte based on the variation in the peak area of a peak in the SES response during the measurement cyclic scan.

4. The method of any one of claims 1 to 3, wherein the discrimination step further comprises discriminating between the first analyte and the second analyte based on the applied potential at which the peak area of a peak in the SES response is at a maximum during the measurement cyclic scan.

5. The method of any one of claims 1 to 4, wherein the discrimination step further comprises discriminating between the first analyte and the second analyte based on the variation of the position of a peak in the SES response during the measurement cyclic scan.

6. The method of any one of claims 1 to 5, wherein the discrimination step further comprises discriminating between the first analyte and the second analyte based on the applied potential at which the position of a peak in the SES response is at a maximum value, and / or the applied potential at which the position of a peak is at a minimum value, during the measurement cyclic scan.

7. The method of any one of claims 1 to 6, wherein the discrimination step further comprises discriminating between the first analyte and the second analyte based on the difference between the SES response when scanning in a positive direction and when scanning in a negative direction during the measurement cyclic scan.

8. The method of any one of claims 1 to 7, wherein the discrimination step further comprises discriminating between the first analyte and the second analyte based on the applied potential at which the peak area of a peak in the SES response is at a maximum when scanning in each of(i) a positive direction, and(ii) a negative direction, during the measurement cyclic scan.

9. The method of any one of claims 1 to 8, wherein the discrimination step further comprises discriminating between the first analyte and the second analyte based on the applied potential at which the position of a peak in the SES response is at a maximum, and / or the applied potential at which the position of the peak is at a minimum, when scanning in each of(i) a positive direction, and(ii) a negative direction, during the measurement cyclic scan.

10. The method of any one of claims 1 to 9, wherein the discrimination step further comprises discriminating between the first analyte and the second analyte based on the variation between a first peak and a second peak in the SES response during the measurement cyclic scan.

11. The method of any one of claims 1 to 10, further comprising a conditioning step prior to the cyclic voltammetry measurement step, the conditioning step comprising a conditioning cyclic scan of potential applied to the SES substrate between a positive potential and a negative potential in the presence of the sample.

12. The method of claim 11 , further comprising a holding step, prior to the conditioning step, in the presence of the sample, wherein a static potential is applied for a period of time.

13. The method of claim 12, wherein the measurement cyclic scan is carried out at an average scan rate of > 5 mV / s.

14. The method of any one of claims 1 to 13, wherein the sample comprises a sensitising electrolyte capable of forming a complex with the first and / or second analyte.

15. A method for in-flow SES analysis, comprising: performing the method of multiplexed detection according to any one of claims 1 to 14, wherein the SES substrate comprises a support and a nanoparticle layer of metallic nanoparticles on the support; then subjecting the nanoparticle layer of the SES substrate to an oxidative cleaning step, thereby producing an oxide coating at the surfaces of the nanoparticles, and subsequently subjecting the nanoparticle layer to a redefinition step, wherein the oxide coating is removed in the presence of scaffolding ligands so that the scaffolding ligands are arranged between adjacent nanoparticles to define their relative spacing.

Images