METHOD, USES OF AND DEVICE FOR SURFACE ENHANCED RAMAN SPECTROSCOPY

Abstract

A method for analysing an analyte (3) using surface enhanced Raman spectroscopy (SERS), comprising the following steps: (a) providing an essentially flat or topologically structured metal surface (1) of a SERS-active metal; (b) depositing the analyte (3) or an open pore matrix material (5) on the surface (1); (c) depositing a multitude of nano-droplets (2) of a SERS-active metal on top of the analyte (3) or the open pore matrix material (5), respectively; and (d) spectroscopically analysing, by scanning laser irradiation and using SERS, the analyte sandwiched between the surface (1) and the multitude of nano-droplets (2). The diameter of the nano-droplets (2) is in the range of 5-70 nm, and the distance between adjacent nano-droplets (2) is smaller than their diameter, and wherein step c) is carried out by PVD or by sputtering SERS-active metal.

Claims

1. A method for analysing an analyte using surface enhanced Raman spectroscopy (SERS), comprising the following steps: a) providing an essentially flat or topologically structured metal surface of a SERS-active metal; b) depositing the analyte or an open pore matrix material on said flat or topologically structured metal surface; c) depositing a multitude of nano-droplets of a SERS-active metal on top of the analyte or the open pore matrix material, respectively; d) spectroscopically analysing, by scanning laser irradiation and using SERS, the analyte sandwiched between the flat or topologically structured metal surface and the multitude of nano-droplets; wherein the number average diameter of the nano-droplets is in the range of 5-70 nm, and wherein the number average distance between adjacent nano-droplets is smaller than their number average diameter, and wherein step c) is carried out by physical vapour deposition (PVD) or by sputtering SERS-active metal; with the proviso that if an open pore matrix material is deposited on the flat or topologically structured metal surface in step b) the analyte is introduced into the pores before step d).

2. The method according to claim 1, wherein the SERS-active metal of the flat or topologically structured metal surface, of the nano-droplets, or of both, is selected from the group consisting of a noble metal, or copper, sodium, potassium or aluminium, or a mixture thereof or an alloy containing such a metal.

3. The method according to claim 1, wherein step b) includes a step of deposing the analyte as a solution, suspension or emulsion in a carrier liquid, and a step of removing the carrier liquid.

4. The method according to claim 1, wherein step c) is carried out by physical vapour deposition (PVD) or by sputtering SERS-active metal, optionally followed by annealing or accompanied by concomitant annealing.

5. The method according to claim 1, wherein in step c) nano-droplets of essentially half spherical shape are generated.

6. The method according to claim 1, wherein the number average diameter of the nano-droplets is in the range of 10-60 nm, and/or wherein the number average distance between adjacent nano-droplets is in the range of 1-30 nm.

7. The method according to claim 1, wherein the thickness of at least one of the analyte layer or of the open pore matrix material sandwiched between the flat or topologically structured metal layer and the nano-droplets is less than 1 m.

8. The method according to claim 1, wherein the flat or topologically structured metal surface is a silver and/or gold layer of a thickness in the range of 5-500 nm, on the substrate material.

9. The method according to claim 1, wherein the analyte is at least one the following: inorganic molecule or particle, organic molecule, including small molecules, DNA molecule, protein, peptide, vitamins, food constituent, cell, including bacterial cells, virus, protozoa, human cells, blood cells, cancer cells, circulating tumour cells.

10. The method according to claim 1, wherein in step d) an irradiation frequency in the range of 600-900 nm is used, and/or wherein in step d) the scanning laser irradiation and using SERS is carried out by way scanning in which the laser and/or the analyte sandwiched between the flat or topologically structured metal surface and the multitude of nano-droplets are moved relative to each other, wherein this scanning can be by way of an X-Y scanning, in which the laser and/or the analyte sandwiched are moved in two directions, or can be by way of the analyte sandwiched being rotated combined with a translational movement of the laser, or can be by way of an oscillating mirror placed under the Raman laser while translating the sample in single axis direction, and/or wherein for the scanning in step d) first, a scanning is performed using a first magnification objective lens for a fast screening and then, scanning is performed using a second magnification objective lens, said second magnification being larger than said first magnification, and/or wherein before, during or after the analysis in step d), the analyte is analysed, using another analytical technique, including XRF, LIBS, or a combination thereof.

11. The method according to claim 1, wherein the flat or topologically structured metal surface comprises the multitude of holes, having a diameter smaller than the analyte particles to be measured.

12. The method according to claim 1, wherein step b) includes a step of fixing the analyte on the flat or topologically structured metal surface and/or wherein, if an open pore matrix material is deposited on the flat or topologically structured surface in step b) the analyte is introduced into the pores before step d).

13. The method of using a method according to claim 1 for solvent quality testing.

14. A multilayer structure for analysing an analyte using surface enhanced Raman spectroscopy (SERS), comprising the following elements: a) an essentially flat or topologically structured metal surface of a SERS-active metal; b) the analyte or an open pore matrix material on said flat or topologically structured metal surface; c) a multitude of nano-droplets of a SERS-active metal deposited on top of the analyte or the open pore matrix material, respectively adapted to be spectroscopically analysed, by scanning laser irradiation and using SERS, the analyte sandwiched between the flat or topologically structured metal surface and the multitude of nano-droplets; wherein the number average diameter of the nano-droplets is in the range of 5-70 nm, and wherein the number average distance between adjacent nano-droplets is smaller than their number average diameter, with the proviso that if an open pore matrix material is deposited on the flat or topologically structured metal surface in step b) the analyte is introduced into the pores before step d).

15. The multilayer structure according to claim 14, wherein the number average diameter of the nano-droplets is in the range of 10-60 nm, and/or wherein the number average distance between adjacent nano-droplets is in the range of 1-30 nm, or wherein the flat or topologically structured metal surface is a silver and/or gold layer of a thickness in the range of 5-500 nm on the substrate material.

16. The method according to claim 1, wherein the SERS-active metal of the flat or topologically structured metal surface, of the nano-droplets, or of both, is selected from the group consisting of a silver, gold, platinum, or copper, sodium, potassium or aluminium, or a mixture thereof or an alloy containing such a metal, and wherein further the method is carried out in one analytical device, in which the actual generation of the flat or topologically structured metal surface of a SERS-active metal according to step a), the analyte deposition according to step b), the deposition of the nano-droplets according to step c) as well as the spectroscopic analysis according to d) take place within one same analytical device.

17. The method according to claim 1, wherein step b) includes a step of deposing the analyte as a solution, suspension or emulsion in a carrier liquid, and a step of removing the carrier liquid, wherein the removal is by evaporation, including by elevated temperature and/or or by reduced pressure, and/or by filtration, and wherein in the latter case the flat or topologically structured metal surface is provided with holes allowing for carrier liquid and non-desired particles to penetrate while keeping analyte particles on the flat or topologically structured metal surface.

18. The method according to claim 1, wherein step c) is carried out by physical vapour deposition (PVD) or by sputtering SERS-active metal, optionally followed by annealing or accompanied by concomitant annealing, and wherein the deposition process is carried out until the layer thickness is not more than 50 nm.

19. The method according to claim 1, wherein step c) is carried out by physical vapour deposition (PVD) or by sputtering SERS-active metal, optionally followed by annealing or accompanied by concomitant annealing, and wherein the deposition process is carried out until the layer thickness is nor more than 15 nm or 9 nm.

20. The method according to claim 1, wherein the number average diameter of the nano-droplets is in the range of 15-50 nm, and/or wherein the number average distance between adjacent nano-droplets is in the range of 5-50 nm.

21. The method according to claim 1, wherein the thickness of the analyte layer and/or of the open pore matrix material sandwiched between the flat or topologically structured metal layer and the nano-droplets is in the range of 1-900 nm.

22. The method according to claim 1, wherein the thickness of the analyte layer and/or of the open pore matrix material sandwiched between the flat or topologically structured metal layer and the nano-droplets is less than 1 m, or in the range of 5-100 nm.

23. The method according to claim 1, wherein the flat or topologically structured metal surface is a silver and/or gold layer of a thickness in the range 10-100 nm, on the substrate material, and wherein the roughness of the surface of the flat or topologically structured metal surface is below 50% of its thickness or the roughness is less than 100 nm.

24. The method according to claim 1, wherein the flat or topologically structured metal surface is a silver and/or gold layer of a thickness in the range 10-100 nm, on the substrate material, and wherein the roughness of the surface of the flat or topologically structured metal surface is below 10% of its thickness or the roughness is less than than 20 nm.

25. The method according to claim 1, wherein the analyte is at least one the following: inorganic molecule or particle, organic molecule, including small molecules, DNA molecule, protein, peptide, vitamins, food constituent, cell, including bacterial cells, virus, protozoa, human cells, blood cells, cancer cells, and circulating tumour cells, and wherein in case of cells also morphological information is determined in the scanning process of step d).

26. The method according to claim 1, wherein in step d) an irradiation frequency in the range of 750-800 nm is used, and wherein further a two-dimensional area is scanned for spectroscopic detection, and/or wherein in step d) the scanning laser irradiation and using SERS is carried out by way scanning in which the laser and/or the analyte sandwiched between the flat or topologically structured metal surface and the multitude of nano-droplets are moved relative to each other, wherein this scanning can be by way of an X-Y scanning, in which the laser and/or the analyte sandwiched are moved in two orthogonal, directions, or can be by way of the analyte sandwiched being rotated combined with a translational movement of the laser, or can be by way of an oscillating, microelectromechanical (MEMS), mirror placed under the Raman laser while translating the sample in single axis direction, and/or wherein for the scanning in step d) first, a scanning is performed using a first magnification objective lens for a fast screening and then, only in the regions providing spectra, scanning is performed using a second magnification objective lens, said second magnification being larger than said first magnification, and/or wherein before, during or after the analysis in step d), the analyte is analysed, in the same device and using the same sample space and preparation, using another analytical technique, including XRF, LIBS, or a combination thereof.

27. The method according to claim 1, wherein the flat or topologically structured metal surface comprises the multitude of holes, having a diameter smaller than the analyte particles to be measured, and wherein the holes have a diameter in the range of 20-200 nm, or in the range of 50-100 nm.

28. The method according to claim 1, wherein step b) includes a step of fixing the analyte on the flat or topologically structured metal surface, by adding a further fixing layer, adding a fixing substance, or by a cross-linking carrier material either deposited together with the analyte on the flat or topologically structured metal surface or before or after deposition of the analyte on the flat or topologically structured metal surface, and/or wherein, if an open pore matrix material is deposited on the flat or topologically structured surface in step b) the analyte is introduced into the pores before step d), after step c), wherein the analyte is introduced into the pores from the gas phase by diffusion or in the liquid phase or as a solution by immersion.

29. The method of use according to claim 13 for water quality testing, including for high purity water testing, in the chip manufacturing field, food and beverage quality testing, pharmaceutical drug discovery, medical diagnostics.

30. The method of use according to claim 13 for water quality testing, using one analytical device in which all steps a)-d) are carried out, and/or analytical device having at least one module for a) generating an essentially flat or topologically structured metal surface of a SERS-active metal; at least one module for b) depositing the analyte or an open pore matrix material on said flat or topologically structured metal surface; at least one module for c) depositing a multitude of nano-droplets of a SERS-active metal on top of the analyte or the open pore matrix material, respectively; at least one module for d) spectroscopically analysing, by scanning laser irradiation and using SERS, the analyte sandwiched between the flat or topologically structured metal surface and the multitude of nano-droplets; wherein the function of the above-mentioned modules can be carried out by individual units or within joint units, and wherein the module for a) and the module for c) are one same unit, which is further an inline device for a supply and which can be fully-automated.

31. The multilayer structure according to claim 14, wherein the number average diameter of the nano-droplets is in the range of 15-15 nm, and/or wherein the number average distance between adjacent nano-droplets is in the range of 5-50 nm, and/or wherein the flat or topologically structured metal surface is a silver and/or gold layer of a thickness in the range of 10-100 nm, on the substrate material, and wherein the roughness of the surface of the flat or topologically structured metal surface is below 50%, of its thickness or the roughness is less than 100 nm, or less than 50 nm.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Preferred embodiments of the invention are described in the following with reference to the drawings, which are for the purpose of illustrating the present preferred embodiments of the invention and not for the purpose of limiting the same. In the drawings,

[0087] FIG. 1 shows a cross section of Ag/analyte/Ag structure: the core technology of U-SERS, wherein in a) the situation of an evaporated or otherwise deposited analyte sandwiched between the Ag elements is given in a cut, in b) the situation where a porous matrix is sandwiched between the Ag elements, the pores of which matrix can take up the analyte for detection in a cut, and in c) a top view onto such a structure.

[0088] FIG. 2 shows a schematic description of U-SERS procedure; the solution to be analysed is drop-casted over a flat Ag surface; after the evaporation of the solvent, a few nm of Ag is deposited creating Ag/analyte/Ag structures shown in FIG. 1; finally Raman scanning is performed over the surface for the detection;

[0089] FIG. 3 shows in (a) distinctive SERS spectra of some pesticides and micropollutants (from bottom to top: BPE, Atenolol, Estradiol, BTAH, Ibuprofen, para-Cresidine, 2-Naphthylamine, 1,2-Dichlorobenzene, 4-Aminophenyl disulphide; in (b) U-SERS map of a 1 ml solution mixture with 30 g Estradiol, 10 g Atenolol, 5 g Ibuprofen, 1 g BTAH and 0.1 g BPE;

[0090] FIG. 4 shows an example of the dimensionality reduction with principle components; a SERS spectrum (left) can be expressed as the linear summation of principle components (three spectra on the right); therefore, the SERS spectrum can be described by only the factors of principle components (1, 2 and 3) instead of thousands of parameters in terms of SERS intensities and Raman wavenumbers;

[0091] FIG. 5 shows a schematic description of main steps of principle component analysis for real-time analysis of chemical mixtures: (1) Determination of the principle components for a specific solution mixture; (2) Determination of the distribution function of analytes as reference library; (3) reduction of the dimension (see FIG. 4); (4) Quantification of the resemblance of the spectra at a certain pixel to the reference library; (5) Quantification of the spectral count;

[0092] FIG. 6 shows a schematic description of the hardware of U-SERS device

[0093] FIG. 7 shows a flat silver substrate can be provided with holes in the sense of a sieve (a), and how the analyte in the form of particles, including bacteria and viruses, can be deposited by allowing a liquid carrying the analyte to pass through the holes keeping the analyte particles on top of the substrate;

[0094] FIG. 8 shows how cross contamination can be avoided by generating the metal surface of a SERS-active metal directly in situ in the analytical device;

[0095] FIG. 9 shows a possible modular set above the analytical device carrying out the proposed method

[0096] FIG. 10 shows another schematic illustration of the structure for analysis on the left side and a SEM picture on the right side;

[0097] FIG. 11 shows another SEM image showing metallic nano-droplet formation on the location of impurities (analytes);

[0098] FIG. 12 shows how the metallic nano-droplet formation makes the deposited impurities on the metallic surface visible;

[0099] FIG. 13 shows an example of a Raman scanning over impurities of ultrapure water after metallic nano-droplet deposition; wherein in (a) a microscope image of the impurities of 300 microliters of ultrapure water on a gold surface after coating the metallic nano-droplets are given; in (b) the area of the Fast-Raman scanning is given, the parameters for the Raman scanning are described in the figure; in (c) the average Raman spectra of the scan is given, this is the average of 10,000 Raman spectra;

[0100] FIG. 14 shows an example of the signal processing to improve signal-to-noise ratio to detect minority impurities;

[0101] FIG. 15 shows the areal intensity;

[0102] FIG. 16 shows the areal intensity versus the impurity content.

DESCRIPTION OF PREFERRED EMBODIMENTS

[0103] U-SERS is based on the Ag/analyte/Ag (Ag: silver) sandwich structure 7 shown in FIG. 1. The structure according to FIG. 1a) comprises a flat Ag thin film 1, a layer of the analyte 3 and Ag nanoislands 2 separated by interspaces 4 can realize a strong electric field enhancement boosting the Raman signals coming from the sandwiched analyte.

[0104] FIG. 1b) shows a corresponding sandwich structure 7 in which the sandwiched layer 5 is not the analyte as such, if needed together with a corresponding binder material, but in which sandwiched layer is given by an open pore material layer. Into this open pore material layer the analyte 3 can be penetrated, by diffusion or by immersion.

[0105] FIG. 1c) shows a top view onto such a structure, showing that the individual nano-droplets 2 are positioned very close to each other and the average distance between the nano-droplets, which typically have a semi-spherical shape and a rather homogeneous diameter of 30, 40 or 50 nm, depending on the manufacturing process, is significantly smaller than the diameter of the nano-droplets 2.

[0106] A possible procedure of making the Ag/analyte/Ag structure is as follows. As shown in FIG. 2, the solution sample is drop-casted on a flat Ag thin-film. Upon drying, the solvent evaporates and the solutes (analytes) are deposited on the Ag surface in random locations. Heat may be supplied to speed up the drying process which can facilitate a more uniform coating of analytes. Then, a few nanometers thick Ag is deposited using sputtering or metal evaporation, forming nanoislands on the analyte layer. Ag nanoisland formation turns the color into a dark tone of blue indicating a strong coupling of light in the red part of the spectra. Indeed, Ag/analyte/Ag structure demonstrates a stronger SERS enhancement for the red excitation laser (785 nm) compared to the green and blue lasers. After the sample preparation procedure, Raman scanning is performed. The distinct blue color of Ag/analyte/Ag structure provides a visual guidance in selecting the areas to be scanned. For the experimental details reference is made to the discussion above.

[0107] FIG. 3 shows the proof of the concept of U-SERS. The spectra of some pesticides and micropollutants are shown in FIG. 3a. It should be noted that only BPE and BTAH among these analytes have a strong affinity to metal. As U-SERS eliminates the effect of the affinity of the molecule to the metallic surface, we can obtain SERS spectra from any analyte as long as it is not highly volatile.

[0108] An important advantage of U-SERS is the ability of the detection of multiple analytes simultaneously. FIG. 3b is an example of such multiplex detection. A solution of five analytes could be resolved easily with a 15-minute-long scan. In the spectral map, each color represents one of the analytes. For the experimental details reference is made to the discussion above.

[0109] The area of a gray shading in the map is correlated with the relative amount of the analyte in the mixture. Even the minority species in the solution (BPE) could be detected at some pixels. Increasing the number of pixels in the map would improve the sensitivity and the quantification accuracy.

[0110] The sample preparation procedure of U-SERS takes about 5 minutes. The scan time, on the other hand, depends on the number of pixels (or the scan area). The state-of-the-art Raman scanners are limited with a scan rate of about 100 spectra/second. Here, it is important to note that 100 spectra/second is already quite fast for ordinary Raman scanning, as the low intensity of Raman signal requires a longer detection time (integration time) per pixel. However, Ag/analyte/Ag structure amplifies the signals about five orders of magnitude indicating a significant room for improvement of the scan rate.

[0111] A U-SERS system should comprise a liquid handling module, a metallization module, and a scanning module.

[0112] State-of-the-art Raman systems perform the data processing after completing the data acquisition. However, post-processing is not a viable strategy, if millions of spectra need to be processed. Target acquisition speed of U-SERS is 10,000 spectra/sec. Such a high scan rate dictates a real-time data processing. Indeed, real-time pattern recognition is a well-established field with various application fields. One example application is the real-time recognition of human face on surveillance cameras. The applicant of this project intends to apply this technology for the real-time processing of SERS spectra. The specific algorithm to be utilized is principle component analysis (PCA). PCA is based on the reduction of the dimension (size) of a spectrum using its principle components. The dimensionality reduction phenomenon is described using a hypothetical example in FIG. 4.

[0113] In order to apply PCA, the principle components need to be determined first. For the U-SERS experiments, this task is a part of the calibration process for a new solution sample. Each of the analytes in the solution needs be described in terms of a statistical distribution function of the principle components. The calibration of the system for a bi-analyte solution can be performed in less than an hour using U-SERS molecular scanner, which is significantly faster than the state-of-the-art chemical analysis techniques such as HPLC where calibration may take weeks. Having the calibration libraries for the analytes, PCA code will be able to process the data real-time. The steps of a typical PCA algorithm are shown in FIG. 5.

[0114] U-SERS contributes to any field requiring a high-performance and affordable chemical analysis technology.

[0115] According to market research, three fields are identified as the market: (1) Food & beverage quality testing, (2) pharmaceutical drug discovery, and (3) medical diagnostics.

[0116] A bottleneck of the food testing is sample purification time. In a typical sample, together with the possible pesticides, a host of other molecules such as vitamins, amino acids and fats would also reside. Such a complex mixture needs to be purified prior to the HPLC test. The purification process can take two to five days. Due to their short shelf lives, fresh foods are introduced on the shelves of the supermarkets before tests results are obtained. The delay of the quality test results clearly imposes risks on the public health. U-SERS can minimize purification procedure due to its multiplex ability. It can even provide comprehensive information about the content of the food (e.g. vitamin, fat content) together with pesticide contamination.

[0117] U-SERS also has a clear potential in pharmaceutical drug discovery where there is a continual effort to develop new HPLC protocols for new solutions. U-SERS with very short calibration time can reduce the overall pharmaceutical drug discovery process significantly. Another area for U-SERS is medical diagnostics. Direct detection of bacteria and virus in the blood is topic for U-SERS applications. Again, the multiplex ability of U-SERS can facilitate the detection of multiple different pathogens while providing additional information about the blood such as the insulin level, combining many different tests into a single U-SERS test.

[0118] As shown in FIG. 6, the hardware is based on a fast and affordable CMOS type spectrometer. As the SERS signal is already amplified through the Ag/analyte/Ag structure, it is possible to compensate the low-efficiency CMOS photodetector. The rotation control of the U-SERS substrate (Ag-coated CD) and the linear movement control of the objective lens is performed via electronic notice board.

[0119] The quantification of analytes is accomplished using two parameters: (i) the number of pixels that a spectrum of a molecule is detected and (ii) the average intensity of certain SERS peaks. At relatively higher concentrations, only the former parameter is sufficient for the quantification.

[0120] During sample preparation procedure, the drop-casting may result in a non-uniform analyte coating known as coffee-ring effect when the solvent evaporation rate is not sufficiently high. In this case, instead of using a flat substrate, a slightly tapered well is used. Such a structure leads the formation of thousands of smaller coffee rings which is sufficient for U-SERS measurement.

[0121] Although the signal enhancement is universe via U-SERS, some of the analytes have intrinsically larger Raman cross sections. When such a Raman-resonant analyte coexists with another one, the signals of the former can dominate the combined spectra. The machine-learning algorithm is constructed considering such challenges.

[0122] FIG. 7 shows how it is possible to provide the substrate with holes 8 of an appropriate size to withhold corresponding particles which one would like to analyse. The particles can be bacteria or viruses, so the substrate can be used as a sieve director, also in a continuous process.

[0123] FIG. 8 shows how very efficiently cross contamination due to surface impurities on the basic substrate, namely the flat or topologically structured surface before the metal is applied, can be avoided. The backing substrate 10, which may contain surface-contamination 11, is directly in the analytical device coated with a clean bottom metallic film 1. Cross-contamination coming from air can be eliminated in that the evaporation process takes place either (i) under vacuum where the pressure is around or larger than 20 mbars, or (ii) under inter gas, e.g. N2 or Argon environment.

[0124] After Raman scanning, one can identify the area of impurities. One can further perform a micro-XRF scanning on the impurities in order to carry out an elemental analysis of the impurities. Raman spectroscopy can be used to identify inorganic and organic impurities. With micro-XRF, it is also possible to detect metallic impurities.

[0125] The actual machine can be composed of different modules, as illustrated in FIG. 9: (1) Sputtering module (2), water-evaporation module, (3) consumables module, (4) Raman scanning module, (5) micro-XRF module and (6) main console for the delivery of consumables between different modules. All of the modules can be under N2 or Argon environment to minimize the cross-contamination.

[0126] FIG. 10 illustrates, how after metal sputtering (or metal evaporation) of layer 1, nano-droplets 2 of metal are formed only on the location of impurities which are the analytes 2 e.g. in case of water analysis as the surface energy of the impurities are lower than metal. We do not form nano-droplets on the locations where there is no impurity (analyte) exist. It should be noted that this process is chemistry-free (not introducing cross-contamination from chemicals). If we had used colloidal nano particles to form such nanoparticles on the impurities, we would introduce impurities of the colloidal solution. In addition, the nanoparticles are dense and uniformly coated on the impurities. The metallic nano-droplet formation is reproducible as it is based on a reproducible process (sputtering or metal-evaporation). In addition, we do not require a chemical bond between the impurities and the nano-droplets. Processes based on colloidal solutions necessitate a chemical bonding between the deposited impurity (analyte) and the nanoparticles. FIG. 11 further gives another SEM image showing metallic nano-droplet formation on the location of impurities. FIG. 12 shows that depositing metallic nano-droplets on the impurities, the metallic nano-droplet-analyte-metal structure absorbs the light. This changes the color on the locations of impurities. So, the analyte location becomes easily visible. This optical effect facilitates a more efficient Raman scanning.

[0127] FIG. 13 gives an example of a Raman scanning over impurities of ultrapure water after metallic nano-droplet deposition. (a) Microscope image of the impurities of 300 microliters of ultrapure water on a gold surface after coating the metallic nano-droplets, wherein 6 nm of Ag is coated with 1 angstrom/s deposition rate. (b) Area of the Fast-Raman scanning. The parameters for the Raman scanning are described in the figure. (c) The average Raman spectrum of the scan. This is the average of 10,000 Raman spectra. In this experiment, total silica, total carbon, and total boron concentration in the ultrapure water were 3000 ppt, 1800 ppt and 13 ppt, respectively. At the average spectrum, one can clearly see the bands defining silica (970 cm1) and carbon (1600 cm1). However, we can also just see the specific band for boron (boric acid) which is at 814 cm1.

[0128] FIG. 14 shows how we implement a signal processing method in order to make the Raman bands of impurities more visible. This signal processing is particularly useful for the detection of minority species (which is boric acid in this experiment). The spectrum (raw data) of FIG. 14 is a sample spectrum of the 10,000 spectra of the scanned map described in FIG. 12. Similar signal processing is performed on each of these 10,000 spectra to identify the peak locations (in terms of Raman shift) and peak intensities. We use the result of this signal processing method in the procedure explained in the following figure.

[0129] FIG. 15 shows, that after the signal processing explained in the previous figure, we can calculate the Areal intensities, which is the multiplication of the total Raman count and total areal coverage of each of the detected Raman bands. FIG. 15a is the same average spectrum as shown in FIG. 13c. FIG. 15b is the transformed spectrum after data processing. The y-axis in FIG. 15a is Raman count. The y-axis of FIG. 15b is areal intensity. Areal intensity is the multiplication of the total intensity and the areal coverage of a certain Raman shift over the scanned area. FIG. 15c is the zoom-in image of FIG. 15b, showing the peak describing Boric acid. The representative U-SERS band for boric acid is 814 cm.sup.1. When we calculate the Areal intensities, we amplify the Raman bands coming from impurities. This amplification is especially helpful to identify the Raman bands coming from minority impurities. In this experiment, total silica, total carbon, and total boron concentration in the ultrapure water were 3000 ppt, 1800 ppt and 13 ppt, respectively. We were easily able to detect boron (boric acid) after signal processing as we could improve the signal-to-noise ratio significantly. It should be noted that the representative U-SERS band for boric acid, 814 cm.sup.1, is hardly visible in FIG. 15a.

[0130] FIG. 16 shows that when we perform the experiment on samples with varying quantities of Boron (boric acid), we can obtain a calibration curve in term of areal intensity of Boron with respect to its weight in the evaporated ultrapure water.

LIST OF REFERENCE SIGNS

[0131]

TABLE-US-00001 1 flat silver substrate, thin film 2 silver nano-droplets 3 analyte .sup.3 deposited analyte to be detected 3 analyte in the pores of 5 4 interspace between 2 5 porous matrix 6 solution mixture 7 Ag/analyte/Ag structure 8 holes in 1 9 particles, viruses/bacteria 10 backing substrate 11 surface-contamination