JP2008519254A - Metal nanovoid photonic crystals for enhanced Raman spectroscopy - Google Patents

Abstract

A planar optical platform for generating a Raman signal from a foreign material includes an input region and an output region for receiving and extracting optical radiation optically coupled to a plasmon band structure region. The plasmon band structure region comprises a layer of a first material having a first index of refraction patterned with an array of subregions of a second material having a second index of refraction, and the sidewalls of each subregion are Coated with a metal dielectric layer. The array of subregions produces a plasmon band structure, and in use, each subregion confines plasmon resonances excited by light radiation coupled within the plasmon band structure region, and the plasmon resonances are in the plasmon band structure region. A Raman signal is generated from a foreign object located in close proximity. The platform may be incorporated into a spectroscopic measurement system and is particularly useful for surface-enhanced Raman spectroscopy of analyte molecules.
[Selection] Figure 2

Description

[Field of the Invention]
The present invention relates to Raman spectroscopy, and more particularly to an optical platform for surface enhanced Raman spectroscopy (SERS).

[Background of the invention]
Raman spectroscopy is used in a variety of applications, most commonly to study vibrational quanta, such as vibrations of molecules or phonons in solids, but other quantized entities are also studied. be able to. Raman spectroscopy can provide detailed information about the physical state of the sample material and is used to distinguish different states of the same molecule, usually different chemically, such as different molecular isomers. Can.

  Raman spectroscopy has a wide range of uses in many different industries. For example, Raman spectroscopy has applications in the fields of pharmaceuticals, chemistry, bioanalysis, medicine, materials science, art restoration, polymers, semiconductors, gemology, forensics, research, military, sensing, and environmental monitoring. .

Raman spectroscopy is a very useful analytical tool, but in practice has a number of drawbacks. A major drawback associated with Raman spectroscopy occurs due to the small scattering cross section. Usually, only 10 ^{-7 of} the photons incident on the sample material will undergo Raman scattering. Therefore, in order to detect Raman scattered photons, the Raman spectrometer usually employs a high-power laser source and a highly sensitive detector. Not only is the scattering cross section small in an absolute sense, but also the scattering cross section is small compared to Rayleigh scattering where the scattered photons have the same energy as the incident photons. This means that there are a number of problems associated with separating small Raman signals from large Rayleigh and incident signals, especially when the Raman signal is close in energy to the incident signal.

  A source of high power is not only bulky and expensive, but at very high power, the intensity of the light radiation itself can destroy the sample material, thus limiting the light source intensity to an upper limit. Provided. Similarly, highly sensitive detectors are often bulky and expensive, especially if forced cooling with liquid nitrogen or the like is required. In addition, it is often a slow process because long integration periods are required to obtain a Raman spectrum signal with an acceptable signal-to-noise ratio (SNR).

  Problems associated with Raman spectroscopy have been known for a long time since C. V. Raman discovered the effect itself in 1928. Since that day, various techniques have been applied to improve the operation of Raman spectrometers.

  Some of the techniques involve the use of metal surfaces to induce surface plasmon resonance (SPR) to more efficiently couple energy to the sample material. One improvement of this technique involves placing the sample material on or near the rough surface. Such a surface can be formed by deposition of metal / dielectric particles, which can be deposited in bulk. A rough surface is found to produce an enhanced Raman signal, and the technique of using a rough surface to obtain a Raman spectrum is known as surface enhanced Raman spectroscopy (SERS).

  However, although SERS devices can provide improved SNR when compared to previous conventional Raman spectrometers, there are still other drawbacks. For example, SERS devices are still not effective enough to provide a Raman signal in the absence of a fairly long detector integration time, and still require a bulky and expensive detector. Even now, it is considered that about 5 seconds is a very good time for collecting Raman spectra. A major problem with many of the currently used SERS systems is the reproducibility of SERS enhancement with respect to the same experimental iterations and between different samples. This is mainly due to the fact that the random distribution and number of hot spots obtained on the rough surface and their location cannot be determined in advance.

  US Patent Application Publication No. 2003/042487 describes a method for providing a metal object selected from nanowires, nanorods, and spheroids. It proposes to surface enhance the spectroscopy of biomolecules very close to metal objects.

  US Patent Application Publication No. 2003/059820 discloses a method, system and method for making a substrate having metal-coated nanoparticles, including nanospheres, nanoparticles, and nanorods (FIG. 7 (b)). See the example shown below).

  US 2003/174384 and US Pat. No. 6,699,724 are based on metal-coated spheres and nanoparticles to enhance Raman signals from analytes in close proximity to the substrate. It explains that it is used as a material.

  US 2002/0668018 discloses a sensor for detecting chemical and / or biological compounds consisting of a plurality of high-Q electromagnetic microcavities. It is claimed that the biological agent is detected by a similar action of SERS.

  US 2004/180379 discloses a SERS-based nanobiosensor, while US Pat. No. 6,579,721 discloses surface plasmon resonance from a perforated metal film placed on a glass substrate. Explains how to biosense using. U.S. Pat. No. 6,759,235 describes a tightly packed, collimated hole distributed over a surface used to immobilize beads for generating a spectrum in response to some external excitation energy. A device comprising a packed array is disclosed.

  Finally, M. Kahl et al. (Phys. Rev. B, vol. 61, no. 20, art. 14078) use a metal-coated 1D grating to enhance the Raman signal from the surface. P. Etchgoin et al. (J. Chem. Phys., Vol. 119, no. 10, p.5281) allow a qualitative understanding of some known SERS phenomena. Therefore, through the analysis of the energy distribution and spatial localization of surface plasmon resonance, the electromagnetic contribution to SERS is detailed from the viewpoint of the photonic crystal concept. Successful results of non-resonant excitation in the electromagnetic contribution to SERS have been explained, and physical phenomena utilizing stimulation or suppression of Stokes / anti-Stokes fields have been proposed. W.L. Barnes et al. (Nature, vol. 424, p. 824) provide an overview of the effects of surface plasmons on various applications.

  Therefore, the role of surface plasmon polaritons (SPP) has been recognized in understanding the heterogeneous transmission properties of light through metal films perforated with subwavelength apertures, and the application of this effect in SERS Was confirmed. However, the combination of the SPP propagating inside the hole, the SPP propagating on the metal interface, and the localized SP constrained by surface defects (and coupling of surface defects) is a Raman signal for use as a biosensor. It was not investigated in terms of enhancing. In addition, the potential benefit of separating light collection and extraction from Raman signal generation, and the possible monolithic integration of signal conditioning functions before and after the Raman generation process, has enabled any investigator in the art. Has not been confirmed.

    [Summary of Invention]

  The present invention provides surface-enhanced Raman spectroscopy that provides reproducible Raman signal enhancement for low concentrations of analyte molecules, as well as additional pre- and post-processing capabilities that can be integrated on a single platform. Platforms, systems, and methods are provided. A platform mass production method is also provided.

  In accordance with a new embodiment presented by the present specification, SERS from low concentrations of analyte molecules incorporated within a PC lattice using metallodielectric photonic crystals (PC). A signal is extracted.

  The present invention encompasses four main aspects: a SERS active platform based on a photonic crystal (PC), a system and method for obtaining a SERS signal from an analyte, and a method for making a PC-based SERS platform. .

  The SERS active platform is based on a coated modulation substrate, whereby an array of holes, polyhedrons, or wells is embedded within the dielectric multilayer system. The dielectric multilayer system can be isotropic in the simplest embodiment, but can also form a distributed Bragg reflector (DBR) or planar waveguide structure. The hole or polyhedron can be disposed on the dielectric multilayer film, but can also partially or fully penetrate the dielectric multilayer film. The holes have a triangular, square, or rectangular lattice geometry, a graded or doubly graded 2D PC lattice, a quasi-periodic 2D PC lattice, or one or more defects. It can be arranged in a regular 2D photonic crystal lattice topology, such as a 2D PC lattice. A polyhedron is an inverted pyramid that normally exists on the surface of a dielectric multilayer system and penetrates into the multilayer (in some cases, a truncated inverted polyhedron). The coating is generally a metal, but can also include other metal or dielectric layers.

  In a preferred embodiment, the metal dielectric layer comprises an optically thin noble metal layer. In another embodiment, the platform is made to be a membrane structure in which the SERS active region is placed over the void. In both the membrane embodiment and the solid substrate embodiment, the metal dielectric layer can be continuous over both the top surface and the sidewall of the hole, or the top surface has no metallization, It is also possible to only cover the hole or the side wall of the polyhedron. The platform may further include one or more functional system blocks, such as a pre-processing unit or a post-processing unit, and / or a sensor unit and a light source unit. The pre-processing unit and the post-processing unit may be infiltrated with an electrically adjustable material, thus increasing the possibility for electrical control of the pre-processing function / post-processing function. Adjacent on-chip system blocks may be combined in an integrated structure.

The system consists of the following seven functional blocks: light source, pre- and post-processing devices, structure for receiving and extracting light radiation, the SERS active platform, and a photodetector system. The functions of the pre-processor can include the selection of incident beam parameters such as the incident polarization angle and azimuth angle, indicated by θ and ψ, the polarization state, and the incident wavelength λ _i shown in FIG. The post-processing unit may, for example, filter out the pump light to improve SNR, introduce wavelength-dependent optical delays, or to spatially resolve the Raman response, to name a few possibilities Super prisms can be included. As described with respect to the platform, some of the functional blocks of the system may be monolithically integrated within a tightly integrated photosensor chip.

  The method describes a means of obtaining a SERS signal for analyte analysis and detection. Analytes are chemical weapons, pesticides, urea, lactic acid, contaminants, ascorbate and glucose, or biomolecules such as proteins, lipids, nucleic acids (bacteria and virus DNA, RNA, PNA, etc.), or cellular substances Can be included. After filling the SERS substrate with the analyte under examination and activating the light source, the SERS signal is detected using a spectrometer. The spectrometer can be a commercially available spectrometer, but may be monolithically integrated on the biochip. The light may be incident out of plane or can be driven into the planar waveguide and is coupled from the planar waveguide to the SERS active region. Similarly, light can be collected out-of-plane or coupled into the underlying waveguide plane and directed to the spectrometer.

  In a final aspect of the invention, a method for mass production of the SERS active platform is provided. Preferably, mass production employs wafer scale silicon compatible fabrication techniques. PC is fabricated by imprinting resist on top of a dielectric substrate such as silicon and then etching cylindrical pores into the substrate using a highly anisotropic etching process Can do. The imprint step allows for inexpensive production and can be easily scaled for high volume production. An alternative means of defining the PC grating structure is a lithographic technique such as photolithography, interference lithography, or electron beam lithography. The SERS active region is created by subsequent deposition of one or more metallo-dielectric layers only inside the pores or both inside the hole and on top of the sample surface.

  In a preferred fabrication method, a lattice pattern definition is applied on the surface of a dielectric substrate consisting of a crystalline silicon wafer oriented along a specific crystal plane. The wafer is then anisotropically etched, typically using a wet etch process. Anisotropic wet etching forms a polyhedron shape, and the exposed polygonal surface of the polyhedron is along the {111} crystal plane of silicon. Thereafter, a metal-dielectric layer application is performed.

  In the preferred orientation of the silicon single crystal wafer, a wafer that is exposed along the {100} crystal plane is selected. An inverted pyramid is formed during the subsequent anisotropic wet etching. When all the polyhedral faces are along the {111} crystal plane of the substrate, the etching process is stopped.

  In another embodiment, the fabrication method is extended to include isotropic deep silicon etching following anisotropic wet etching. Thereby, a hole with a high aspect ratio can be arranged at the center of the polyhedron.

  In a further embodiment of the present invention, a plasmon band structure is provided so as to provide a means to maintain a highly reproducible SERS signal even when there are rough corrugations on the surface of the metal dielectric photonic crystal. An area may be designed. The rough waveform results from the fabrication deposition method and varies randomly. It is known in the prior art that surface roughness is one contributing source of SERS signal generation. However, the randomness of roughness generates local hot spots where SERS is formed, and thus large variations in SERS signals that occur at different locations on the SERS substrate (> 100% of the standard deviation of SERS signals) May cause. The present invention maintains plasmons and localized plasmons that are substantially related to the large-scale features of plasmon band structure patterning, rather than the small-scale features of nanoscale surface roughness (<100 nm). High reproducibility is achieved by optimizing the band structure characteristics.

  In yet another embodiment of the invention, the plasmon band structure region is designed to have a repeating array of subregions, each subregion having a plurality of inverted polyhedrons or holes of different sizes and shapes. Each size and shape provides a different localized plasmon resonance, thus enabling increased coupling efficiency for input laser light of any wavelength and enhanced output coupling for Raman light of different wavelengths. . This provides improved reproducibility of the output SERS signal.

  The main improvements provided by the present invention compared to the prior art include the following aspects. The topology allows light injection from the side, thus facilitating an easier and more reproducible means of light injection into the SERS active region. Second, the integration of many functional system blocks into a highly integrated chip further reduces the complexity of the system because of the reduced number of external components required, and thus System costs will be reduced. Smaller system configurations will be further evaluated, for example, in in vivo endoscopy. Third, utilizing the separation of light injection, Raman signal generation, and light extraction, each functional block can be optimized for its specific role (eg, conflicting with maximum SERS enhancement). To extract light as it does).

  Finally, compared to colloidal crystal-dependent topologies, the proposed fabrication method that defines patterns by lithographic means promises higher reproducibility and increased yield, in contrast to self-organized methods . The scalability of the novel anisotropic wet etch fabrication process and deep silicon etch fabrication process for the production of SERS substrates provides a cost-effective way to mass produce the same SERS substrates while providing different experiments, the same It still maintains high signal reproducibility across different substrates on the wafer and / or across different substrates on different wafers.

  The present invention will now be described in detail with reference to the accompanying drawings.

Detailed Description of the Invention
SERS Active Platform An important aspect of the present invention is a planar platform for receiving light radiation from a light source, as described herein.

  In the first embodiment of the platform shown in FIG. 1 (a), the platform comprises a dielectric layer having a predefined surface modulation (patterning) coated with a continuous metal dielectric coating 102, 105, 106. I have.

  Surface modulation is achieved by the definition of a two-dimensional (2D) periodic array of holes or recesses 101 incorporated into a normally flat dielectric surface 103. The topology of the 2D periodic grating structure can include, but is not limited to, square, triangular, and rectangular grating geometries. The array of holes can also be graded 2D photonic crystal lattice, doubly graded 2D photonic crystal lattice, quasi-periodic photonic crystal, or one or more Can be arranged to form a lattice structure having a number of defects. If the array of holes is arranged in a regular grid, it will be characterized by a unique grid pitch. In a preferred embodiment, the holes or recesses 202 have the shape of cylindrical air rods having a hole diameter 206 and a height 203 as shown in FIG.

  In another embodiment, the hole or recess 110 has the shape of an inverted pyramid (FIG. 1 (e)) or an inverted truncated pyramid 111 (FIG. 1 (g)) having a square base length 115 and a height 116. The recess is not limited to an inverted pyramid and may take the form of an inverted polyhedron.

  In another preferred embodiment (FIGS. 1 (i) and 1 (k)), the holes or recesses have a hybrid shape (top view 118) that combines an inverted pyramid and a deeply extended cylindrical pore. It may be taken. A high aspect ratio pore extends from the top of an inverted pyramid into the underlying substrate.

  The structure is coated with a metal layer or a metal dielectric layer that can include several metal films and a dielectric film. In the simplest embodiment, the layer is a single metal film. Each layer labeled j can be of a different material and generally has a thickness 204 at the top 102 of the structure, a thickness 205 at the hole (recess) sidewall 105, and a thickness 207 at the hole (recess) floor 106. Will be different. In a preferred embodiment, the thickness of the metal layer is selected to be optically thin. The metal can be selected from, but not limited to, the following groups of metals: gold, platinum, silver, copper, palladium, cobalt, nickel, and iron.

  In order to improve the adhesion of the metal to the dielectric surface, an adhesive layer 208 such as a single layer of chromium or mercapto-propyltrimethoxysilane ("mercaptosilane") can be applied to any dielectric material as shown in FIG. On top, it can be deposited before depositing the metal layer.

  In one exemplary embodiment, the modulated dielectric surface is coated with a continuous metal film. The continuous metal film covers the top surface and the sidewalls and bottom of the hole as shown in FIG. In another embodiment shown in FIG. 1 (b), only the holes (recesses) are coated with metal films 105 and 106, and there is no metal film on the top surface of the structure.

  In yet another embodiment, when the metal covers both the top surface and the sidewalls, the metal coated dielectric is made to take a film structure as shown in FIG. 1 (c). In this configuration, the metal-coated dielectric grating is often referred to as an air bridge or film structure because it is cut off by the air region 109. The PC region is mechanically connected by a dielectric region 107 surrounding the hole 101. The deposited metal dielectric layer will also be selectively deposited at the bottom of the air cavity. As shown schematically in FIG. 11, in this membrane configuration, the analyte can flow through the platform 1101 from one end 1102 to the other end 1103 during measurement, and the light radiation 1104 is incident from the top, for example, The Raman signal 1105 emanating from is detected, for example, in the reflection regime. This configuration gives the experimenter the freedom to mechanically control the adsorption properties (chemical binding energy, adsorption efficiency, etc.) of the analyte molecules on the SERS active region. Optically, any waveguide mode that propagates in the core of the structure is symmetrically affected by the cladding and buffer material (ie, air) and can therefore set up a symmetric mode. This can provide a means to optically probe and excite analyte molecules 211 located above and below the air bridge structure. FIG. 1 (d) shows yet another embodiment similar to the previous configuration, in which only selected areas of the hole sidewall 105 and the bottom of the cavity are coated with a metal dielectric layer.

  In the case of FIG. 1 (e), FIG. 1 (g), and FIG. 1 (i), the continuous metal film 117 has an inverted pyramid (FIG. 1 (e)), an inverted truncated pyramid (FIG. 1 (g)), And the angled facets 112 in the inverted pyramidal rod hybrid structure (the high aspect ratio structure shown in FIG. 1 (i)) and the bottom surface 113 (if present).

  In other embodiments shown in FIGS. 1 (f), 1 (h), and 1 (k), an inverted pyramid (FIG. 1 (f)), an inverted truncated pyramid (FIG. 1 (h)), and inverted. Only the pyramidal rod hybrid structure (FIG. 1 (k)) is coated with the metal film. As shown in FIG. 1 (f), FIG. 1 (h), and FIG. 1 (k), the metal film 117 exists only along the inclined facets (FIG. 1 (e), It is not present on the top surface of the structure (compared to FIG. 1 (g) and FIG. 1 (i)).

  Further, in all embodiments, the substrate consists of a multilayer film 104 of dielectric material, and in its simplest embodiment consists of a single dielectric layer. In the example of FIGS. 1E to 1K, the single dielectric layer is not limited, but is made of a single crystal substrate such as silicon, and the inclined facet 112 is formed along the crystal plane. The In order to provide a simplified manufacturing method in which an anisotropic polyhedron can form an inverted polyhedron by following the crystal plane, a single crystal substrate is required.

Other embodiments of the present invention include a planar waveguide structure 212 covered by the substrate 108 and the cladding 103 shown in FIG. 2, and a distributed Bragg reflector 310 shown in FIG. is there. The dielectric layer may have a residual distance 210 to the bottom of the hole (recess), or may be partially or completely perforated by the hole (recess). In the preferred embodiment of the planar waveguide 212, its thickness 209 and the distance 210 from the floor of the hole (recess) provide optimal coupling to the SPP while still allowing the light of the preselected wavelength λ _i to be vertical. Selected so that it can be trapped in. In a preferred embodiment of a planar distributed Bragg reflector (DBR) mirror, the thickness of the layers forming the mirror is selected so that the center of the stop band coincides with the preselected wavelength λ _i of the incident light. Is done. In another embodiment, the DBR mirror can be designed such that the reflective region of the structure matches the most important wavelength region of the generated Raman signal and can provide a high SNR. The multilayer structure can also be configured to form an omnidirectional reflector (ODR).

The wavelength λ _{i of the} incident light is the plasma frequency of the surface metal ω _p = (4πnq ² / m ^* ) ^1/2 (1)
When the frequency is close to, conduction electrons on the metal surface can be excited to a surface electron excited state called surface plasmon that will be mainly localized to defects on the metal surface. Here, n represents an electron density, q represents an electronic charge, and m ^* represents an effective mass. When there is no periodic modulation on the metal surface, the dispersion relation of the surface plasmon is
k _SP = k ₀ (ε _D ε _M / (ε _D + ε _M )) ^1/2 (2)
Given in. Where λ _M is the frequency dependent complex permittivity of the metal, λ _D is the frequency dependent complex permittivity of the dielectric, k _SP is the complex wave number vector of the surface plasmon, and k ₀ is the incident It is a wave vector of light emission.

Light excites localized surface plasmons (SP) at defects in the structure, but if certain conditions are met, such as the proposed topology of the present invention, the air-metal interface It is also possible to excite surface plasmons polaritons (SPP). That is, by introducing a periodic modulation of the complex permittivity of the metal, the angle of the incident light is changed to the following phase matching condition:
k _SPP = ± k ₀ sin θ ± mQ (3)
Choose to be satisfied. Here, Q = π / Λ is a reciprocal lattice vector of a metal surface periodically corrugated in a specific lattice direction with a corrugation period Λ.

The modulation period Λ is equal to half the wavelength λ _SPP of the surface plasmon polariton, ie
Λ = λ _SPP / 2 (4)
In this case, the periodic wave function of the surface plasmon polariton can form a standing wave on the surface of the metal. Then, the SPP bandgap that does not allow the presence of the surface plasmon polariton state is formed, resulting in disturbance of the SPP dispersion (2). This phenomenon is equivalent to formation of a photonic band gap in a photonic band gap structure or formation of a band gap for electrons in a solid.

  Similarly, the SPP can be excited inside the hole and the hole can combine with the SPP on the top surface.

The electric field (E _r E _θ E _z ) and magnetic field (H _r H _θ H _z ) solved by the cylindrical coordinates (r, θ, z) for the surface plasmon dispersion in a single metal cylinder of finite thickness by the following formula: Is shown:

であり、ｊ番目の層の波数ベクトルの半径方向成分は、ｋ_ｊ ^２＝（２π／λ）（ｎ_ｊ ^２−ｎ_{ｅｆｆｅｃｔｉｖｅ} ^２）^１／２であり、ｎ_{ｅｆｆｅｃｔｉｖｅ}は、プラズモンモードの有効指数である。 Where ε _j = ε ₀ n _j ² , ε _j is the dielectric constant of medium j, n _j is the refractive index of medium j, and ε ₀ is the dielectric constant of vacuum,
Exponential factor

And the radial component of the wave vector of the j-th layer is k _j ² = (2π / λ) (n _j ² −n _effective ² ) ^1/2 , and n _effective is an effective exponent of the plasmon mode. is there.

  The cylindrical structure consists of three layers (j = 3), and the first radial medium is the cylindrical core region (j = 0), usually air. The second medium consists of metal (j = 1), and finally the third medium (j = 3) consists of a dielectric material, i.e. a cladding structure.

The general solution for this set of equations consists of I and K form Bessel functions, but with appropriate boundary conditions, ie, the tangential components of E and H at the cylinder inner and outer diameters of the metal film. The condition that it is continuous must also apply. Matching of the electromagnetic field components at the boundary results in a system of linear equations for the coefficients a ^j _n and b ^j _n . Non-trivial transcendental equations will have roots that give surface plasmon dispersion relations (derived from n _effective ). This equation is solved numerically by methods such as the bisection method and Newton-Raphson method, or preferably by a more accurate method such as the Augmented Principle Method.

  In addition, when the holes are very close to each other, the evanescent field extending into the dielectric can couple from one hole to a nearby hole, thus reducing the band gap even in the absence of a connected metal top surface. Band structures that can be shown are formed independently.

  When modeling the SP mode in an inverted pyramid structure, the auxiliary differential equation-finite difference time domain (ADE-FDTD) method is used.

  The dielectric constant (ε) of the metal layer is approximated by the use of a sixth-order Lorentz function, as shown below.

Where j is the number of oscillators and each oscillator has an intensity of f _j and a frequency ω _j , while ω _p is the metal plasma frequency.

  A normal incidence TE polarized Gaussian packet is reflected from the surface of the structure and the reflectance power versus wavelength is collected. An example TE polarization simulation is plotted in FIG. 14 where reflected power 1401 is plotted against wavelength 1402. The geometry of the simulated structure is shown highlighted in FIG. 14B. The pyramid periodicity 1409 is selected to be 2 mm, while the pyramid size 1405 varies from 0.5 mm to 1.5 mm, the angle 1407 is selected as 45 °, and the thickness of the selected gold metal The length 1404 is 0.3 mm. Some resonances are visible on the plot. Note that the region 1410 where the drop in reflection occurs indicates the region where the surface plasmons are coupled, thereby preventing light from being reflected. These are the most important areas, and the structure is designed so that the SPP resonant coupling matches the desired input wavelength.

  FIG. 18A shows an experimental reflectance spectrum over a range of a plasmon band structure substrate of the type shown in FIG. 17A. Normalized reflectivity for normal incidence is plotted against the wavelength indicated in nanometers. The pitch in the plasmon band structure substrate is fixed at 2 microns, while the inverted pyramid diameter varies from 0.9325 μm to 0.7825 μm. Over this range, the plasmon resonance appears to shift to higher energy (shown as a wavelength shift 1801 from long wavelength 1802 to short wavelength 1803). This is theoretically verified using 2D ADE-FDTD in 18B. That is, when the inverted pyramid diameter changes from 1.25 μm to 1.00 μm, the plasmon resonance also shifts blue 1806, causing a resonance wavelength shift from 1807 to 1808.

  Furthermore, as shown in FIG. 18C, plasmon energy is obtained in volt units from FDTD simulation results (shown as dashed line 1812) and experimental results (shown as medium black circle 1811). Plotted against reciprocal of hole depth. The second-order (m = 2) and third-order (m = 3) modes in both theory and experiment are highlighted, both having the same energy spacing and the same dependence on pit depth Have sex. The visible shift between experiment and theory is that the experiment is carried out on a 3D inverted pyramid, whereas the theory is that the 2D simulation is a surface on a V-groove type structure (which is the effective cross section of an inverted pyramid structure). By modeling plasmons.

  Alternative simple and intuitive models have also been proposed. This is based on confinement of propagating surface plasmon polaritons (SPPs) to the sides of an inverted pyramid. This model assumes that the top of the inverted polyhedron serves as an infinite reflection barrier for surface plasmons, and the sharp inverted hole bottom conveys surface plasmons. Using these assumptions, a simple model can be derived that predicts resonant plasmon energy as follows.

は、換算プランク定数（reduced Planck’s constant）である。 Where n _SPP is the effective refractive index of the SPP mode on flat gold, ω is the angular frequency of the incident light, m is the order of the plasmon mode, and d is the inverted pyramid depth. And α is the angle of inclination of the pyramid surface with respect to the vertical (in this case 35.3 degrees), while

Is a reduced Planck's constant.

  It can be seen that the first two modes of the simple model are superimposed on FIG. 18C (open square 1813), which is in good agreement with the plasmon resonance obtained from the FDTD model. Recognize. FIG. 18D shows the localization of the modeled TM polarization field in the inverted pyramid structure and the associated surface and localized plasmons. FIG. 18E depicts an intuitively predicted field distribution based on a simple model in which propagating surface plasmons are confined within an inverted pyramid. Localized plasmons can thus be approximated by delocalized surface plasmons that move up and down the side wall of the inverted hole. This provides a plasmon mode that can form a resonant standing wave between nodes at the upper edge of the hole.

The ADE-FDTD method is also used to localize along the surface of the PC-SERS structure the time-average square of the incident laser pump field <E ² _LASER > and the localized time-average square of the Raman field <E ² _RAMAN > can be quantified. Use this data to calculate the effective figure of merit (FOM) for the SERS enhancement level due to field localization by defining FOM = <E ² _LASER ><E ² _RAMAN > May be. Using FOM provides a means to predict SERS enhancement and, as a result, predict control of reproducibility across the surface.

  It is known in the prior art that small scale roughness (on the order of about 10 nm to 100 nm) can play a fundamental role in the generation of SERS. However, roughness is inherent in most metal deposition processes. For example, depending on the fabrication process selected, the metal dielectric layer can have a random waveform of metal dielectric micro-columnar shapes. This microcolumnar roughness can also be modeled using the ADE-FDTD method. This technique provides a way to model the localization properties and SERS FOM associated with both surface roughness and PC-SERS structure.

  FIG. 15A shows a localization simulation of a time-averaged TM polarized laser pump light with a wavelength of 785 nm incident on an inverted pyramid PC-SERS structure having a smooth surface. The structure includes a silicon oxide substrate 1514, a deposited metal layer 1513 having a thickness of 300 nm, and an upper air region 1515. Results are calculated for a structure in which inverted pyramids having a pyramid size of 650 nm are arranged in a square array of 1 micron pitch. Strong localized plasmons 1512 and weak surface plasmons 1511 are clearly visible.

  The corresponding results for a flat but corrugated gold metal surface are shown in FIG. 15B. This corrugated flat gold metal structure is a first order approximation as a nanoscale periodic structure of air column cavities having a roughness pitch of 70 nm, a diameter of 50 nm and a depth of 100 nm. Modeled. The microcolumnar structure can provide a large SERS FOM by localizing strong Mie-like plasmons around a single microcolumnar in roughness, as shown at 1516. Are known.

  FIG. 15C shows the electromagnetic field localization obtained when the coarse waveform is superimposed on the PC-SERS structure of FIG. 15A. The localization of the 785 nm incident TM polarized laser light is mainly in the rough waveform (small-scale Mie-like resonance) 1517, but partly because of surface plasmons and localized plasmons (due to the large-scale plasmon band structure). Form. However, in a preferred embodiment, the photonic crystal SERS is such that surface plasmons and localized plasmons (1518) are dominant, which are independent of small coarse waveforms but depend on large plasmon band structure features. Region parameters and physical properties of the metal layer are selected. A simulation result for the structure optimized in this way is shown in FIG. 15D. In the particular example shown, the depth of the micro-columnar rough corrugations varies from 25 nm depth in the structure of FIG. 15C (providing large small-scale Mie-like plasmon resonance) to 100 nm depth in the structure of FIG. 15D. This type of optimized design ensures that high reproducibility is obtained from the SERS substrate, regardless of the waveform roughness characteristics.

  Photonic crystal lattice parameters can be selected to extend the lifetime of surface plasmons, increase the chance of binding to analyte molecules, and thus increase measurement sensitivity. Furthermore, the angular direction of the emitted Raman signal may be changed by appropriate changes in the array pattern.

  When the analyte 211 is brought very close to the metal dielectric layer, energy is coupled to the analyte molecules from various plasmon states, exciting various vibration modes, resulting in an enhancement of the Raman signal. The very large EM field associated with SPP can enhance molecular vibration modes, primarily vibration modes perpendicular to the metal surface. The molecule relaxes and returns energy to a plasmonic surface state that emits EM radiation or mediates the emission of a Raman signal.

  The SERS enhancement depends mainly on the distance of the molecule from the metal surface. Two mechanisms contribute to SERS enhancement. That is, electromagnetic (EM) contribution and chemical contribution.

  Depending on the distance of the adsorbate molecules to the metal surface, the EM or chemical contribution becomes more pronounced. In general, the closer the bond of the adsorbate to the metal to form the adsorbate-molecule complex, the stronger the chemical action, although the action of EM is still dominant.

  In addition, the structure provides a means to adjust detection to be sensitive to adsorbates further away from the metal dielectric surface by changing the depth of penetration into the evanescent field air cavities. It can also be designed to provide. This may be achieved by changing the photonic crystal lattice parameters and the thickness of the metal dielectric layer. The resulting structure can be beneficial for the detection of large analyte molecules and biological material such as cells, bacteria, or viruses where large characteristic dimensions will be detected.

  As described above, the remote contribution is considered to be due to resonance excitation of the SPP by the incident light wave. However, when the adsorbate is in contact with the metal surface, it forms a metal-adsorbate complex so that the proximity contribution to enhancement can no longer be ignored. Proximity contributions are likely due to changes in the ground state electronic charge density distribution (as a result of charge transfer (CT) excitons) and new electronic excitation spectra.

  So far, the charge transfer (CT) mechanism is not fully understood. Two main approaches are possible: a CT mechanism similar to the resonant Raman mechanism and a non-radioactive CT mechanism. Depending on the adsorption strength of the molecules on the metal surface, either of the two models appears to dominate the close-up enhancement mechanism.

  Analyte molecules are in close proximity to the metal at the bottom or sidewall of the hole, the metal at the opening of the hole, the defect area in the PC lattice, or the metal dielectric coated top surface of the structure, if applicable. Can be in a state.

  Regardless of the specific mechanism details, the essential role of SPP in enhancing the vibrational modes of molecules adsorbed on metal surfaces is clear. Of course, it becomes clear that the redistribution of the SPP state is likely to increase the SERS effect basically.

The energy difference between the incident signal and the Raman scattered light is equivalent to the vibrational energy of the adsorbate molecule, and the Raman shift Δk
_Δk = 1 / λ _i + 1 / λ _scatter (8)
Can be expressed as

  The Raman shift is characteristic of the adsorbate molecule because it depends on the vibrational mode of the adsorbate molecule and can therefore function as a fingerprint of the analyte under examination. Raman shifted signals are collected out of the plane of the platform or in the plane of the waveguide structure.

  Analytes include chemical weapons, pesticides, urea, lactic acid, contaminants, ascorbate and glucose, or biomolecules such as proteins, lipids, nucleic acids (bacterial and viral DNA, RNA, PNA, etc.) Can do.

  The platform has an area for receiving light radiation and an area for extraction. These regions can be selected from, but not limited to, the group of metal photonic band structures, metal-coated dielectric photonic band structures, or dielectric photonic band structures listed below. Diffraction gratings, graded gratings, periodic two-dimensional photonic crystals, graded two-dimensional photonic crystals, or graded or doubly graded two-dimensional photonic crystals, quasi-periodic 2 It can be selected from the group of dimensional photonic crystals. Regular features can include, but are not limited to, square, triangular, or rectangular geometries. The filling rate of air rods or air slots, and the respective etching depths, can transport light to or from the platform that generates the Raman signal. Can be selected to facilitate optimal coupling of light to or from the in-plane wave vector of the resulting multilayer structure. An exemplary embodiment of such a structure is shown in FIG.

  FIG. 4A shows a diffraction grating structure 401 having grating parameters (grating pitch, groove width, groove depth) selected so that light is optimally coupled to the in-plane wave vector of the underlying multilayer film structure. Indicates. FIG. 4 (b) shows a second exemplary embodiment for receiving or retrieving EM radiation, where a 2D photonic band structure device 404 with pores arranged in a rectangular array is shown. Show. FIG. 4 (c) shows a photonic crystal 405 having a quasi-periodic grating topology that can be used to generate a modified emission pattern of light emission. FIG. 4 (d) shows a fourth exemplary embodiment of a structure for receiving or extracting light radiation, in this case showing the cleaved facet 408 of the device. A rib or ridge structure 407 can be directed to the edge of the facet. The waveguide can be further tapered so that it is mode matched with an external optical fiber for EM wave implantation (402) or collection (403).

  FIG. 4 (a) shows a preferred embodiment for realizing a structure for receiving light radiation, while FIG. 4 (c) shows a preferred embodiment of a structure for extracting light radiation.

  The platform can further include a structure 503 for pre-processing and / or post-processing of light radiation. In a preferred embodiment, the recess completely penetrates the planar waveguide structure located below the platform surface, as shown in FIG.

  The topology of the photonic band structure (grating period, air hole diameter and depth) allows the structure to have a preselected function such that an incident optical signal having intensity 501 at wavelength 502 is subjected to some form of optical processing. Selected to meet. For example, the following processes: spectral filtering, polarization filtering, optical delay, beam direction control, diffraction, and spatial resolution of multiple different photon energies. If the pores are infiltrated with an adjustable material such as a nematic liquid crystal or an electrically modulated nonlinear material, and electrodes are applied to the top and bottom of the substrate, the respective functions can be easily changed electronically. For example, changing the applied voltage results in a polarization dependent shift of the photonic band gap of the photonic crystal (PC) portion. This can function as a polarization filter that can be electronically controlled by blocking or passing light in a specific polarization state, or electronically controlled by allowing or not allowing light in a certain frequency range to pass. It can be used to function as a possible wavelength filter.

  The waveguide structure 212 can transport energy from one functional block 601, 602, 604, 605 to an adjacent functional block. The refractive index of the stack, the thickness 209 of the waveguide core, and the distance to the top surface (the sum of distances 203 and 210) are selected to minimize out-of-plane radiation loss. This is accomplished by solving the waveguide dispersion relationship and selecting a mode that does not introduce any loss. Furthermore, the waveguide mode can be confined laterally using one of several mechanisms. An example is an optical mode provided by a photonic crystal line defect, such as a coupled resonator optical waveguide (CROW), also known as a coupled-cavity waveguide (CCW). Resonant coupling. Another example is effective refractive index guiding provided by a ridge or rib waveguide, or a wide photonic crystal channel waveguide with more than two defect rows. It is. Depending on the waveguide mechanism, a mode and phase matching structure at the end of the waveguide can be defined to improve coupling between adjacent system blocks.

  FIG. 6 shows three exemplary embodiments of an integrated platform. FIG. 6 (a) shows a structure 601 that receives light radiation (from left to right), eg a preprocessing unit 602 for selecting a preferred polarization state, a platform 604 that generates a Raman signal, eg after filtering out pump light. FIG. 6 shows a platform incorporating a processing unit 605 and a cleaved facet 606 that extracts the Raman signal. The dielectric multilayer structure located below the SERS active region forms a planar waveguide structure 212 that carries light from one functional block to another. The light 402 is incident on a structure 601 designed to receive light radiation, the in-plane wave vector component of the incident light is coupled to the planar waveguide 212, and the incident light is based on the PC by the planar waveguide 212. To the pre-processing portion 602. After passing through the preprocessing unit 602, the light has a defined polarization state. This polarized light binds to the SPP, which excites an oscillation mode of the analyte molecule 603 that is located very close to the active SERS region 604. The Raman signal exiting the SERS active region recombines into the waveguide 212 and passes through a post-processing unit 605 that filters out the wavelength of the incident light radiation 402, for example, and is directed to the cleaved facet 606 therefrom. The light 403 can be extracted.

  FIG. 6 (b) shows a monolithic integrated device in which the structure for receiving and extracting light radiation and the active structure for generating a Raman signal are integrated into one photonic crystal portion 607. Has been. Incident light 402 in a predefined polarization state is coupled out of plane to the SERS active region of the platform. Utilizing photonic band structure characteristics and coupling light into the desired Bloch modes and at the same time interacting with SP mode to allow several operations to be performed simultaneously Can do. The Raman enhanced electromagnetic signal 403 exiting the platform is collected from the same plane from which the platform received radiation. That is, the measurement is performed in a reflection regime. The polarization state of incident light can be defined by passing light through an external polarizing filter. In this embodiment, the platform preferably comprises a multilayer mirror structure 301 located below the photonic crystal SERS active surface, and its stop band overlaps with the desired Raman signal, thereby reflecting the Raman signal. And guide into the reflection regime. This configuration represents a preferred embodiment for the SERS active platform.

  FIG. 6 (c) shows a third embodiment in which the cleaved facet 606 functions as a structure that receives light radiation. Incident electromagnetic radiation 402 is guided from the input facet 606 through the planar waveguide structure 212 and coupled to the photonic band structure, which is coupled to the SP mode of the SERS active region. The spacing between the photonic band structure and the waveguide determines the coupling strength. That is, the shorter the distance, the stronger the bond. The SERS active region and the structure for extracting light radiation are integrated into one structural unit 608. The Raman signal 403 exiting the SERS active region is collected out of plane for post-processing and detection.

System In a second aspect of the present invention, a system for analyte detection and analysis is provided, as schematically illustrated in FIG. The system can use less incident radiation and still provide a Raman enhancement signal such that an acceptable signal-to-noise ratio (SNR) is obtained compared to standard Raman measurements.

  The system includes a light source 1001 that emits electromagnetic radiation, a functional unit 1002 that pre-processes the optical signal, a structure 1003 that receives the optical radiation, a structure 1006 that generates a Raman signal, and a functional unit 1004 that post-processes the Raman enhancement signal. And a structure 1005 for extracting light radiation and a sensor 1007 for detecting Raman-enhanced electromagnetic radiation. The unit 1002 for pre-processing light radiation, the unit 1003 for receiving light radiation, and the unit 1004 for post-processing light radiation and the unit 1005 for extracting light radiation are replaced by alternative branches as shown schematically in FIG. can do. That is, the functional pre-processing / post-processing unit can be integrated within the device or external to the device.

  The preprocessing functions shown in FIG. 10 can include polarization control and / or filtering of incident EM radiation, which can be accomplished either externally with individual polarizers and optical filters, or in-plane. When coupling light in-plane, the pre-processing can be monolithically integrated on a single chip with the SERS active platform, as previously described in the platform. For example, the PC portion can be defined in-plane to act as a wavelength filter or a polarizing filter.

  The post-processing unit can filter the pump light to improve the SNR, filter the desired polarization state for detection, or by spatially resolving the different energies of the emitted photons It can be used to act as a monolithically integrated spectrum analyzer. If the Raman signal is detected in-plane, the Raman signal can be directed to a separate PC portion that, for example, penetrates the waveguide core and serves as a super prism, polarizing filter, or spectral filter. Alternatively, the PC grating that constitutes the SERS active region can be defined such that the propagation direction of the SPP is very sensitive to the respective energy.

  Therefore, it is possible to confirm two types of component integration that can independently bring down the size of the system equipment and the reduction in the number of necessary components constituting the system. First, the functional block may be integrated on the same chip with the SERS active region. Secondly, adjacent structural blocks defined on the same biochip may be combined into one structural element.

  An example of the first integration type is shown in FIG. By defining an active region located adjacent to the passive waveguide region, active components other than those already mentioned in the system, i.e., a light source emitting light radiation and a sensor detecting the SERS signal, are placed on the biochip. So that it is not necessary to provide an external light source and / or an external photodetector.

  Examples of the second integration type are shown in FIGS. 6 (b) and 6 (c). Adjacent to each other, functional blocks for receiving and / or extracting optical radiation and functional blocks for generating Raman signals are integrated in one structural element 607 and 608, respectively.

  Accordingly, many of the functional blocks shown in FIG. 10 can be monolithically integrated into a monolithic integrated device, thereby reducing the number of individual components that form the system, and thus reducing space and component costs. Saved.

Method In a third aspect of the invention, a method is provided for detecting and analyzing an analyte using the platform and system described above. This method is shown schematically in FIG. Analyte molecules are brought close to the PC SERS active region (851). The light source is turned on (852) and the electromagnetic radiation passes through the pretreatment unit. The setting of the pre-processing unit allows light of a predefined incident parameter such as wavelength, polarization state and / or angle of incidence to be applied to a planar waveguide structure located in a region above or below the surface. From either, it is adjusted to bind to the SERS active region (853).

  Light excites various plasmon states. Here, the various plasmon states can be bonded to each other, and energy can be transferred to a plurality of different vibration modes of the adsorbate molecule. The Raman signal is collected and passed through a post-processing unit and detected either in-plane or out-of-plane depending on the selected system configuration using a spectrometer or other means outlined above ( 854).

  FIGS. 16A, 16B, and 16C show the experimental Raman signals detected using the claimed PC-SERS substrate in counts versus wave numbers. A self-assembled monolayer (SAM) of benzenethiol is placed on the surface of the PC-SERS substrate and detected using a Renishaw Raman spectrometer. SEM micrographs of the PC-SERS structure are shown inserted in FIGS. 16A, 16B, and 16C. All structural metal coatings are nominally 300 nm.

  In the structure shown in FIG. 16A, the SERS active region comprises a metal coated square lattice pore photonic crystal region. The lattice pitch of the pores is about 600 nm, a scale that is almost an order of magnitude larger than prior art structures related to SERS enhancement with a rough waveform (˜50 nm). As shown, the Raman signal from the benzenethiol SAM highlights the feasibility of detecting SERS from the PC-SERS substrate. Because of their small dimensions, these structures are usually lithographically defined using expensive direct write electron beam lithography, or alternatively, nanoimprinting.

For more cost effective fabrication, the structure shown in FIG. 16B consists of a metal-coated modulated photonic crystal surface with inverted pyramids arranged at a 2 μm pitch, while the structure shown in FIG. 16C is an inverted pyramid. And comprises a metal-coated corrugated surface with a deep etched cylindrical hole located at the apex of an inverted pyramid. These structures can be defined using low cost photolithography. Deep (high aspect ratio) cylindrical holes are typically deeper than 40 μm, and in this example have a gold metal coating of 0.3 μm at a 6 μm pitch. To highlight the excellent reproducibility, several Raman signals for benzenethiol SAM were collected at 25 different locations around the 25 μm × 25 μm area of the PC-SERS substrate and superimposed on the same plot. Has been. The relative standard deviation of peak height at 1071 cm ⁻¹ relative to the background level for the different experiments was calculated as 7.0% for the high aspect ratio structure and 9.9% for the inverted pyramid structure. It should also be noted that the Raman signal background (characterized by a wide band background signal detected over a large wavenumber range) can be significantly reduced depending on the PC-SERS structure used. In particular, the high aspect ratio structure shown in FIG. 16C dramatically reduces the effect. This is due to the efficient coupling of the Raman signal out of plane with directivity, as opposed to the omnidirectional scattered background signal.

  To characterize the possible contribution to the SERS signal resulting from the surface roughness waveform in the PC-SERS active region, a metal-coated inverted pyramid SERS structure high-resolution field emission electron gun SEM (FEG-SEM) is shown in FIG. As shown in FIG. It can be seen that the roughness of the sidewall is about 70 nm, and about 20 nm air cavities surround the micropillars having a depth of 50 nm to 100 nm.

  For a control comparison, the SERS effect of benzenethiol SAM was investigated on a flat unpatterned gold layer with the same rough waveform. The measured SERS signal obtained is shown in FIG. 17B (shown with surface SEM inserted). The dotted line shows the Raman spectrum acquired using the same acquisition time and settings used in the experiments that resulted in FIGS. 16A, 16B, and 16C, while the solid line is 6 times longer Shows time. The apparent lack of Raman signatures means that benzenethiol is not detected and therefore the flat rough gold surface under test is not a SERS active substrate.

  However, a close examination of the dispersion characteristics of the metal-coated inverted pyramid SERS structure of FIG. 17A reveals many prominent features due to the long scale patterned plasmon band structure. FIG. 17C shows the results of a reflectance experiment in which collimated laser beams of various wavelengths are incident on the structure at different angles of incidence. Plasmon energy (and hence angular frequency) is plotted against incident angle θ. The resulting contour map shows a 2D plasmon dispersion diagram of the inverted pyramid PC-SERS substrate. Several thinned lines 1706 show the coupling of incident light into several different diffraction modes associated with large-scale periodic modulation of the structure. However, the discrete and complex pattern of islands 1703 arises from surface plasmons and localized plasmons and is attributed to the metal-coated inverted pyramid structure. SERS can be achieved when light is coupled into these localized plasmon regions. During the selection of the PC-SERS parameters, the placement and richness of these islands can be adjusted to optimize the SERS effect, resulting in a highly reproducible SERS signal.

  Experiments were conducted on structures where the PC pitch varied from 0.4 μm to 6 μm, and the structures showed excellent reproducibility with less than 15% variation from sample to sample.

In FIG. 18F, the effect on the SERS signal by matching localized plasmon resonance to the pump laser wavelength is evident. In this experiment, a monolayer coating of 4-aminothiophenol is attached to a tilted inverted pyramid substrate similar to the substrate described with reference to FIG. 18A. From the plot of FIG. 18F, the SERS signal (count unit) for the 1,080 cm ⁻¹ line resonates with the excited pump laser light by changing the inverted pit depth in micrometer units. It can clearly be seen that it is significantly enhanced as it enters the state.

  In another aspect of the present invention, FIG. 19 is a plan view of a metal dielectric photonic crystal structure that is not limited and that is tolerant in fabrication and is not wavelength sensitive, comprising an array of small regions of inverted pyramids. Indicates. The small areas defined by 1901 and 1902 in the horizontal and vertical directions are composed of a plurality of inverted polyhedrons of different sizes arranged at the same pitch 1903. Each inverted polyhedron in the small region provides a different localized plasmon resonance, thus allowing optimal field localization for a range of operating input laser wavelengths. Furthermore, the localized plasmon resonance in that range allows output Raman light of a plurality of different wavelengths to be recombined with the plasmon resonance and propagated to the output coupling region. In order to improve the reproducibility of the SERS signal, the incident laser spot size 1904 is set to overlap at least one small region.

Method of Making In a fourth aspect of the invention, a method of making a SERS active platform is provided. The manufacturing process is schematically shown in FIGS. 8A (a) to 8A (k), and includes an epitaxial growth step, a pattern defining step, a highly anisotropic etching process, and a metal deposition step.

  In the first step shown in FIG. 8A (b), the multilayer dielectric structure 104 is epitaxially grown on top of the substrate 108 (FIG. 8A (a)) to form, for example, a multilayer mirror structure or waveguide structure. . Epitaxial growth techniques include, but are not limited to, the following groups of techniques: molecular beam epitaxy (MBE), metalorganic vapor phase epitaxy (MOVPE), gas source molecular beam epitaxy (GSMBE) / chemistry Line Epitaxy (CBE) / Metal Organic Molecular Beam Epitaxy (MOMBE), Liquid Phase Epitaxy (LPE), Atomic Layer Epitaxy (ALE), Hydride (or Halide) Vapor Phase Epitaxy (HVPE), Vapor Phase Growth, Solid Phase Epitaxy You can choose from.

  In the second step shown in FIG. 8A (c), an etch mask layer 802 is preferably deposited on the organically cleaned surface of the cladding 103, followed by a resist layer 801. The etch mask layer 802 can be composed of a dielectric material or a metal material, and its thickness is selected to serve as a shadow mask in a subsequent etch process until the desired etch depth is reached. The mask material can be deposited using, for example, vacuum sputtering or an electron beam evaporation process.

  The pattern can then be defined in the resist by using one of the following methods: photolithography, electron beam lithography, interference lithography, or imprint lithography, followed by resist development. (FIG. 8A (d)). The pattern is then transferred to the etch mask using a highly selective etching process (FIG. 8A (e)). Here, this highly selective etching process selectively removes the etching mask 802 but does not etch into the upper layer material 103.

  Using the second highly anisotropic etching process, the hole or recess 804 is transferred to the substrate using the perforated layer 803 as a shadow etch mask as shown in FIG. 8A (f). This process includes the following exemplary etching processes: electron cyclotron assisted reactive ion etching (ECR-RIE), chemically assisted ion beam etching. ) (CAIBE), inductive coupled plasma reactive ion etching (ICP-RIE), and ion beam milling or anodic etching can be used.

  Depending on whether it is desirable to have a metal coating on top of the substrate surface, the remainder of the etch mask is removed or retained, respectively. As shown schematically in FIG. 8A (j), if removal of the etch mask is desired, it can be accomplished using either a dry etch process or a wet etch process.

  An adhesion layer 208, such as a monolayer of chromium or mercapto-propyltrimethoxysilane, is deposited on the metal surface as shown in FIGS. 8A (g) and 8A (j) to bond the noble metal to the substrate. Is improved. The noble metal is then deposited or plated (electroplating and electroless plating) on top of the adhesive layer on top of the sample 103 and inside the holes 105 and 106 (FIG. 8A (h) And FIG. 8A (k)). To ensure that the structure has a metal dielectric layer at the bottom and sides of the pores, the particle flow 1201 is tilted by the appropriate angle shown in FIG. 12 with respect to the normal of the substrate surface. The structure 1203 on which the substrate 1202 is mounted is continuously rotated while material is deposited on the sample. If the etch mask has not been previously removed, it is now removed to form isolated and metal coated 105, 106 air holes (FIG. 8A (i)).

  Depending on the method of metal deposition or plating, the rough waveform of the metal dielectric layer can be controlled. These are usually in the small roughness range of about 10 nm to 100 nm. Roughness can be controlled by fabrication deposition process conditions that can affect the degree and size of roughness. Examples of relevant parameters are deposition rate, substrate temperature, r. f. Includes power level, chamber vacuum pressure. The depth of the rough corrugation can also depend on the thickness of the deposited metal dielectric layer.

  The manufactured structure has a metal dielectric surface shape modulation schematically shown in FIG. That is, the upper surface 1301 of the metal dielectric coated hole is lower than the upper surface 1302 of the underlying dielectric structure.

  Another fabrication method employs anisotropic wet etching that defines the holes / recesses following the resist pattern definition. Anisotropic wet etching is a known technique and is commonly used in the MEM area. Selective etching along specific faces of single crystal silicon is employed to define a predetermined inverted polyhedral structure on the surface of the substrate. Its shape is determined by selecting the crystal orientation of the silicon wafer, the resist pattern, and the etching duration. Once all the inverted polyhedron surfaces are along the {111} crystal plane of the substrate, the etching process is temporarily stopped. In the simplest form, a {100} single crystal wafer surface 850 is selected as shown in FIG. 8B (a). An etch mask layer 851 is preferably deposited on the organically cleaned surface of single crystal silicon. The etching mask layer 851 can be made of a dielectric material that is selectively resistant to anisotropic wet etching. These are usually silicon nitride or silicon dioxide. Thereafter, using one of the previously described techniques, a pattern is defined in resist 852 and then transferred into the mask (FIG. 8B (b)).

Using an anisotropic wet etching process, a recess 854 is formed as shown in FIGS. 8B (c) and 8B (g). Examples of chemical wet etch process used in the anisotropic wet etching, EDP (ethylenediamine pyrocatechol), CsOhH, a KOH, NaOH, or _N 2 _H 4 _-H 2 0 (hydrazine). Note also that the etch rate can also be controlled by temperature and concentration. In the case of a {100} surface oriented wafer, an inverted pyramid structure is formed if etching is allowed to stop spontaneously. The angle formed by the upper surface of the substrate (shown in broken lines in FIGS. 8B (c) and 8B (g)) and the inclined facet is about 54 °. The position of the inverted pyramid is determined by the position of the etched small area in the mask. However, it should be noted that if the anisotropic wet etch is allowed to stop spontaneously, the shape of the array elements on the mask becomes meaningless and a structure such as FIG. 8B (g) is formed. The shape of the inverted polyhedron to be formed is finally determined by the orientation of the silicon crystal plane. In the case of {100}, an inverted pyramid is formed. The size of the inverted pyramid depends on the size of the mask subregion 857 and will expand until the periphery of the small region is completely enclosed by the pyramid base 856 as shown in FIG. 8B (k). If anisotropic wet etching is allowed to finish early, a more complex truncated pyramid structure is formed. This is shown in FIG. 8B (c) for {100} crystal substrate orientation.

  In another preferred embodiment, an additional etching step is further performed following the anisotropic wet etching. This provides a structure that combines an inverted pyramid and a hole as shown in FIG. 8B (l). The cross section of the hole 858 takes the shape of a small area of the etching mask (as in 857). Etching can be performed using highly anisotropic reactive ion etching, ion beam milling, or anodic etching. A structure with a hole depth versus lattice pitch greater than 10: 1 can be achieved.

  As stated previously, depending on whether a metal coating on the top substrate surface is required, the remainder of the etch mask is replaced by FIGS. 8B (d) and 8B (f) and FIG. h) and removed or retained, as shown in FIG. 8B (j). As schematically shown in FIGS. 8B (e) and 8B (i), when etching mask removal is desired, either dry or wet etching processes are used to achieve etching mask removal. be able to.

  Following the definition of the inverted polyhedron and the high aspect ratio hole, the steps of metal layer deposition are the same as those previously described in connection with FIG. 8A.

1 (a)-(k) illustrate ten exemplary embodiments of the present invention. (FIG. 1 (a) shows a photonic crystal (PC) area showing a continuous metal dielectric film on its surface and recess. FIG. 1 (c) shows that the SERS active region is an air cavity. 1 (b) and 1 (d) correspond to FIGS. 1 (a) and 1 (c), respectively, where only the recesses are coated with a metal dielectric layer. The structure is shown. FIG. 1 (e) shows another embodiment of the present invention showing a photonic crystal (PC) having an inverted pyramid recess, and FIG. 1 (g) shows a truncated inverted pyramid recess. FIG. 1 (f) and FIG. 1 (h) show an inverted pyramid and a truncated inverted pyramid in which only the recesses are metal-coated. 1 (i) and 1 (k) have an inverted pyramid structure and a cylindrical hole extending into the substrate from the apex of the pyramid, each having a metal coating continuous throughout the structure, and Fig. 3 shows a structure in which the metal coating is present only in the recess. ) FIG. 6 illustrates an exemplary embodiment disclosed in the present invention, in which a multilayer dielectric structure having a PC SERS region defined therein forms a waveguide structure. FIG. 6 illustrates an embodiment in which a multilayer dielectric structure having a PC SERS region defined therein forms a multilayer mirror structure. FIG. 4 shows four exemplary embodiments of structures designed to receive or extract light radiation. 1 is a schematic representation of a structure designed to pre-process or post-process light radiation. FIG. 3 shows three exemplary embodiments, one embodiment in which one or more functional system units are integrated on a chip (a), two combined one or more functional system units An embodiment (b) in which a new structure incorporating the above functions is formed, and an embodiment (c) in which the same structure remains. 2 shows two examples of prior art systems. FIG. FIG. 6 illustrates a fabrication process disclosed by the present invention for creating a SERS active platform. 8B (a) -8B (j) illustrate the fabrication process disclosed by the present invention when producing a SERS active platform using anisotropic etching of a single crystal silicon wafer. On the other hand, FIG. 8B (k) is a diagram showing the anisotropic wet etching characteristics of an inverted pyramid when etched through a mask. FIG. 8B (l) shows a structure resulting from the combination of anisotropic wet etching and highly anisotropic anode etching. 1 is a schematic diagram illustrating a method for obtaining a Raman signal from an analyte under test. 1 is a schematic diagram illustrating a system for obtaining a Raman signal from an analyte under test. FIG. 6 is a schematic diagram illustrating a setup in which an analyte flows continuously through the platform during a measurement. FIG. 4 shows a deposition process of a metal dielectric layer. 1 is a schematic diagram showing a fabricated SERS structure. It is a figure which shows the theoretical reflectance simulation result about the surface of the metal-coated inverted pyramid PC-SERS structure. FIG. 14B shows the characteristics and geometry of the metal-coated inverted pyramid PC-SERS structure in FIG. 14A. Figures 15A, B, C and D relate to the effect of the surface waveform. FIG. 15A is a diagram showing simulation results of TM field localization and related surface plasmons and localized plasmons in the structure of FIG. 14B having a smooth surface. It is a figure which shows the localization of TM field relevant to the rough waveform formed in the flat surface. It is a figure which shows TM field localization in the inverted pyramid which formed the rough surface waveform by metal deposition. FIG. 4 is a diagram showing TM field localization in an optimally designed pyramid having a rough waveform. FIGS. 16A, B, and C show from this benzenethiol self-assembled monolayer when the benzenethiol self-assembled monolayer was placed on the surface of each of the three types of PC-SERS substrates. It is a figure which shows the reproducibility of the experimental Raman signal obtained (it shows by inserting a scanning electron microscope result (SEM)). FIG. 16A is experimental Raman data from a metal-coated air rod PC-SERS substrate. The result of the experiment SERS from the inverted pyramid PC-SERS base material is shown. It is a figure which shows the reproducibility of the Raman signal from the PC-SERS base material which has a hole with a deep etching (a high aspect ratio). Figures 17A, B and C relate to the effect of the surface waveform. FIG. 17A is a photomicrograph of the high-resolution field emission electron gun SEM of the inverted pyramid PC-SERS structure of FIG. 16B and shows the surface roughness. It is a figure which shows the experimental SERS signal (The characteristic Raman line is lacking) obtained from the benzenethiol monolayer provided on the flat metal substrate which has a rough waveform (Inserted SEM shows surface roughness) ). It is a figure which shows the 2D plot of the dispersion | distribution characteristic measured about the inverted pyramid PC-SERS structure. FIG. 18A is an experimental reflectivity plot at normal incidence showing plasmon resonance variability in structures with different pyramid sizes. FIG. 18B is a theoretically modeled reflectance plot at normal incidence showing plasmon resonance variability in structures with different pyramid sizes but fixed pitches. FIG. 18C plots the energy for the first few of the confined surface plasmon (SP) states against the pit depth. Results for a simple theoretical model, full Finite-Difference Time-Domain (FDTD) modeling, and actual experiments are shown. FIG. 18D is a diagram showing TM polarization field localization in an inverted pyramid structure, and surface plasmons and localized plasmons associated therewith. FIG. 18E is a diagram showing a predicted field distribution based on a simple model in which propagating surface plasmons are confined within an inverted pyramid. FIG. 18F shows that the SERS signal obtained from a monolayer coating of 4-aminothiophenol varies as a function of the diameter of the inverted pyramid. A metal dielectric photonic crystal structure that is tolerant in fabrication and not wavelength sensitive, designed to have many different localized and broad surface plasmon resonances, and a SERS incident laser spot Metal dielectric photonic crystal structure that provides more efficient input laser coupling and more reproducible SERS output signal regardless of operating wavelength by allowing size to cover many unit elements It is a top view.

Claims (39)

An input region for receiving light radiation;
A plasmon band structure region optically coupled to the input region, comprising a layer of a first material patterned with an array of small regions of a second material, the first material having a first refractive index Wherein the second material has a second refractive index, and the sidewalls of each subregion are coated with a metal layer or metal dielectric layer, the array of subregions resulting in a plasmon band structure, In use, each subregion confines plasmon resonances excited by light radiation coupled within the plasmon band structure region, which is from foreign material placed in close proximity to the plasmon band structure region. A plasmon band structure region that produces a Raman signal;
A planar optical platform for generating a Raman signal from a foreign material, comprising an output region optically coupled to the plasmon band structure region for extracting light radiation.   The platform of claim 1, wherein the end walls of each subregion are coated with a metal dielectric layer.   3. A platform according to claim 1 or claim 2, wherein the unpatterned region of the layer of the first material is coated with a metal dielectric layer.   The small region includes a square, rectangular, or triangular geometric lattice, quasi-periodic tiling, amorphous tiling, graded lattice, doubly-graded lattice, and two tiling arrays. Claim 1-Claim characterized in that it is arranged at the apex of a predefined tiling array selected from the group comprising superposition and tiling arrays having single or multiple defective subregions. 4. The platform according to any one of 3.   The at least one subregion comprises a column or elliptical cylinder having a diameter from 5 nm to about 10,000 nm and a cylindrical depth from 1 nm to 10,000 nm. The platform according to any one of the above.   The at least one subregion comprises an inverted polyhedron or an inverted truncated polyhedron having a base length in the range of 50 nm to about 20,000 nm. platform.   The at least one small region includes an inverted polyhedron or an inverted truncated polyhedron continuous with a cylinder surface, an elliptic cylinder surface, or a cylinder having a polygonal surface, and the maximum diameter of the cylinder is smaller than the polyhedron base length. The platform according to claim 6.   The platform according to claim 6, wherein the polyhedron is formed of a pyramid.   9. A platform according to any one of the preceding claims, wherein the transverse extent of each of the subregions varies across the array.   The platform according to claim 1, wherein the second material is made of air.   11. The first material of claim 1, wherein the first material is selected from the group comprising silicon, silicon dioxide, silicon nitride, silicon oxynitride, tantalum pentoxide, polymer, and semiconductor material. The platform according to item 1.   The platform according to claim 1, wherein the metal layer or the metal dielectric layer has a thickness of 1 nm to 500 nm.   The metal layer or the metal dielectric layer includes one or more metal layers including a metal selected from the group including gold, platinum, silver, copper, palladium, cobalt, iron, and nickel. The platform according to any one of claims 1 to 12.   14. The platform according to claim 1, wherein a surface of the metal layer or the metal dielectric layer is rough because of a coating manufacturing method.   The platform according to any one of claims 1 to 14, further comprising a multilayer planar structure made of a metal dielectric or a dielectric on which the patterned layer is disposed.   The platform of claim 15, wherein the multilayer structure comprises one of a planar distributed Bragg reflector, a planar optical waveguide, and a one-dimensional photonic crystal stack.   17. The platform of claim 15 or 16, wherein the multilayer structure comprises a material selected from the group comprising silicon, silicon dioxide, silicon nitride, silicon oxynitride, tantalum pentoxide, polymer, and semiconductor material. .   The platform according to any one of claims 15 to 17, wherein the small region extends through a part of the multilayer structure.   The platform according to any one of claims 15 to 18, wherein the multilayer structure includes a gap immediately below the small region.   The platform according to any one of claims 16 to 19, wherein the input region includes a part of the planar optical waveguide.   20. The input region comprises a portion of the plasmon band structure region, and in use, diffractively couples light radiation into the plasmon band structure region. Platform described in.   A platform according to any one of claims 1 to 21 and the plasmon band for pre-processing light radiation incident on the platform and coupling the pre-processed radiation to the plasmon band structure region. A processing structure optically coupled to a structure region, or a processing structure optically coupled to the plasmon band structure region for post-processing light radiation coupled from the plasmon band structure region to the post-processing structure; An optical device comprising:   The processing structure performs one or more functions selected from the group including polarization filtering, spectral filtering, optical time delay, super-diffraction, beam direction control, and beam collimation. The optical device according to claim 22.   The processing structure is selected from the group comprising a taper coupler, an evanescent coupler, a velocity taper coupler, a grating structure, a one-dimensional photonic crystal, a periodic two-dimensional photonic crystal, and a two-dimensional photonic quasicrystal. Item 23. The optical device according to Item 22.   25. The optical device according to claim 22, wherein the processing structure is formed integrally with the plasmon band structure region. A light source;
26. The optical platform of any one of claims 1-21, or the optical device of any one of claims 22-25, optically coupled to the light source;
To detect a Raman signal generated when the foreign object is placed in proximity to the plasmon band structure region optically coupled to the optical platform or the optical device and to which radiation from the light source is coupled An optical system for performing Raman spectroscopy on a foreign object. The optical platform according to any one of claims 1 to 21 or the optical platform of the optical device according to any one of claims 22 to 25 in proximity to the plasmon band structure region. Placing sample foreign matter;
Activating a light source that generates light radiation optimized for excitation of surface plasmons, wherein the optimized parameters for radiation include a group comprising wavelength, azimuth and polarization angle, and polarization state A step selected from
Coupling the light radiation into the plasmon band structure region;
Detecting a spectrum of a Raman signal emanating from the optical platform, and obtaining a Raman spectrum or a surface enhanced Raman spectrum from a sample foreign material.   29. The method of claim 28, wherein the step of disposing the foreign material includes flowing the foreign material into a fluid near the plasmon band structure region.   The foreign substance is an analyte molecule selected from a biochemical group including chemical weapons, agricultural chemicals, urea, lactic acid, contaminants, ascorbate, glucose, or nucleic acids such as bacterial and viral DNA, RNA, PNA, etc. 29. A method according to claim 27 or claim 28 comprising protein, lipid, and a biomolecule selected from the group comprising biological cells such as cancer cells.   30. The method of claim 29, wherein the extraneous matter is disposed within 50 nm of one of the subregions within the plasmon band structure region. Epitaxially growing the layer of the first material and an optional underlying multilayer structure;
Defining a pattern on the layer of the first material;
Etching a portion of the epitaxial structure with anisotropic etching;
22. A method of making an optical platform according to any one of claims 1-21, comprising depositing a metal layer or a metal dielectric layer.   32. The method of claim 31, wherein the pattern defining step is selected from the group of techniques including photolithography, deep UV lithography, electron beam lithography, interference lithography, imprint process, or stamping process.   The etching steps include reactive ion etching (RIE), electron cyclotron resonance reactive ion etching (ECR-RIE), reactive gas assisted ion beam etching (CAIBE), and inductively coupled plasma reactive ion etching (ICP-). 33. A method according to claim 31 or claim 32, comprising a highly anisotropic plasma etch performed using a technique selected from the group comprising (RIE).   33. A method according to claim 31 or claim 32, wherein the etching step comprises highly anisotropic anodic etching.   33. A method according to claim 31 or claim 32, wherein the etching step is highly anisotropic and includes ion beam milling. The anisotropic etching is wet etching using a chemical selected from the group comprising EDP (ethylenediamine pyrocatechal), CsOhH, NaOH, N ₂ H ₄ —H ₂₀ (hydrazine), and KOH. 33. A method according to claim 31 or claim 32, wherein the method is a process.   The first material is single crystal silicon, and the anisotropic etching is terminated when the {111} crystal plane of silicon is completely exposed along all the faces of the inverted polyhedron. The method of claim 36.   36. A method according to any one of claims 31 to 35, further comprising an additional etching step according to claim 36 or claim 37. 32. The metal deposition step is performed using a technique selected from the group comprising physical vapor deposition (PVD), vapor deposition, chemical vapor deposition (CVD), and atomic layer deposition (ALD). The method according to any one of claims 38 to 38.

Images