US20060060472A1

US20060060472A1 - Microstructures and method of manufacture

Info

Publication number: US20060060472A1
Application number: US11/229,772
Authority: US
Inventors: Tadabumi Tomita; Yoshinori Hotta; Akio Uesugi
Original assignee: Fuji Photo Film Co Ltd
Current assignee: Fujifilm Holdings Corp; Fujifilm Corp
Priority date: 2004-09-22
Filing date: 2005-09-20
Publication date: 2006-03-23
Also published as: JP4395038B2; JP2006089788A

Abstract

In order to provide a structure which has a sufficiently large signal strength and generates local plasmon resonance of excellent reproducibility, the invention provides a process for manufacturing a structure at least a portion of which is an aluminum member having on a surface thereof an anodized layer having a plurality of micropores, the process comprising the steps of, in order, anodizing a surface of an aluminum member so as to form an anodized layer having micropores present therein, sealing the micropores in the anodized layer by filling the micropores with metal, surface-treating the sealed anodized layer so as to remove at least a portion of upper layer surface thereof and set the average surface roughness (R_a) to at most 30 nm, and subjecting the surface-treated anodized layer to electrodeposition so as to form metal particles on the metal filled into the micropores during sealing.

Description

The entire contents of literatures cited in this specification are incorporated herein by reference.

BACKGROUND OF THE INVENTION

The present invention relates to a microstructure which uses an aluminum member having on a surface thereof an anodized layer having a plurality of micropores. The invention also relates to a process for manufacturing such a microstructure.
Within technical fields having to do with, for example, metal and semiconductor thin films, fine lines and dots, it is known that the movement of free electrons becomes confined at sizes smaller than some characteristic length, as a result of which peculiar electrical, optical and chemical phenomena become observable. Such phenomena are called “quantum-mechanical size effects” or simply “quantum size effects.” Active research and development is currently being conducted on functional materials which employ such phenomena. Specifically, materials having structures smaller than several hundred nanometers in size are called microstructures or nanostructures, and are today one target of material development efforts.
Methods for manufacturing such microstructures include processes in which a nanostructure is directly manufactured by semiconductor fabrication technology, including micropatterning technology such as photolithography, electron beam lithography and x-ray lithography.
Of particular note is the large amount of research being conducted today on processes for manufacturing nanostructures having an orderly microstructure.
An example of a method of forming an orderly structure in a self-ordering manner is an anodized alumina film (anodized layer) obtained by anodizing aluminum in an electrolyte solution. It is known that a plurality of micropores having diameters of about several nanometers to several hundreds of nanometers form in a regular arrangement in the anodized layer. It is also known that by using the self-ordering nature of this anodized layer to obtain a perfectly regular arrangement, theoretically there will form hexagonal columnar cells wherein the base is a regular hexagon centered on a micropore and the lines connecting neighboring micropores form equilateral triangles.
For example, H. Masuda et al. (Jpn. J. Appl. Phys. Vol. 37, Part 2, No. 11A, L1340-1342 (Nov. 1, 1998), FIG. 2) describes an anodized layer having micropores. In another related publication, Hy{overscore (o)}men Gijitsu Binran [Handbook of surface technology], edited by The Surface Finishing Society of Japan (1998)(Nikkan Kogyo Shimbun, Ltd.), pp. 490-553, the natural formation of micropores as oxidation proceeds in an anodized layer is described. Moreover, H. Masuda (“Highly ordered metal nanohole array based on anodized alumina,” Kotai Butsuri Vol. 31, No. 5, 493-499 (1996)) has even proposed the formation of a gold dot array on a silicon substrate using a porous anodized layer as the mask.
The greatest characteristic of an anodized layer as a material is held to be that the plurality of micropores take on a honeycomb structure in which they are parallelly formed in a direction that is substantially vertical to the substrate surface, and at substantially equal intervals. In addition, the ability to relatively freely control the pore diameter, pore spacing and pore depth is also regarded as a characteristic lacking in other materials (see Masuda, 1996).
Known examples of applications for anodized layers include various types of devices, such as nanodevices, magnetic devices and light-emitting devices. For example, JP 2000-31462 A describes such applications as magnetic devices in which the micropores are filled with the magnetic metal cobalt or nickel, light-emitting devices in which the micropores are filled with the light-emitting material ZnO, and biosensors in which the micropores are filled with enzymes/antibodies.
In addition, in the field of biosensing, JP 2003-268592 A describes an example in which a structure obtained by filling the interior of micropores in an anodized layer with metal is used as a sample holder for Raman spectroscopy.
Raman scattering is the effect where, when incident light (photons) strikes particles and scatters, non-elastic collisions with the particles arise, causing a change in energy. Raman scattering is used as a technique for spectroscopic analysis, but a current challenge is how to enhance the intensity of the scattered light used in measurement so as to improve the sensitivity and accuracy of analysis.
A known phenomenon that enhances Raman scattered light is the surface-enhanced resonance Raman scattering (SERRS) effect. This effect is one where the scattering of certain kinds of molecules absorbed onto the surface of, for example, a metal electrode, a sol, a crystal, a vapor-deposited film or a semiconductor, is enhanced relative to within a solution. A remarkable enhancement effect of from 10¹¹to 10¹⁴times is seen particularly in gold and silver. The mechanism underlying the SERRS effect is not yet fully understood, although the surface plasmon resonance described below is believed to play a role. Use of the plasmon resonance principle as a means for enhancing the Raman scattering intensity is a stated object in JP 2003-268592 A as well.
Plasmon resonance is the effect where, when the surface of a noble metal such as gold or silver is irradiated with light so that the metal surface is placed in an excited state, plasmon waves—which are the localized electron density waves, interact with electromagnetic waves (resonance excitation) to form a resonance state. Surface plasmon resonance (SPR) is a type of plasmon resonance in which, when the metal surface is irradiated with light, free electrons at the metal surface acquire an excited state and collectively oscillate, generating a surface plasmon wave which in turn generates a strong electric field.
In the near-surface region where plasmon resonance arises, that is, in the region within about 200 nm from the surface, an electric field enhancement of several decades (in one case, 10⁸to 10¹⁰times) can be seen, and a distinct rise is observed in various optical effects. For example, when light is directed at a prism having thereon a vapor-deposited thin film of a suitable metal such as gold at an angle larger than the critical angle, changes in the dielectric constant of the thin-film surface can be detected to a high sensitivity as changes in the intensity of the reflected light due to the surface plasmon resonance effect.
Specifically, using a SPR sensor which employs the surface plasmon resonance effect, quantitative measurements of reactions and bonds between biomolecules and kinetic analysis can be carried out without labeling and in real time. SPR sensors are used in research on immune response, signal transmission, and interactions between various substances such as proteins and nucleic acids. Recently, a paper was even published on analyzing trace dioxins using a SPR sensor (Karube, et al., Analytica Chimica Acta Vol. 434, No. 2, 223-230 (2001)).
Various methods are being studied for increasing plasmon resonance, including techniques that involve localizing plasmons by using the metal in the form of discrete particles rather than as a thin film. For example, JP 2003-268592 A describes a technique in which localization is induced by providing metal particles on regularly arranged pores in an anodized layer.
According to a recent research article, when local plasmon resonance with metal particles is used, if the metal particles are present in close proximity to each other, there is achieved a state wherein the electric field strength is enhanced in the gaps between the metal particles, making it easier to generate a plasmon resonance (see T. Okamoto: “A study on metal nanoparticle interactions and biosensors,” found in an Internet search on Nov. 27, 2003 at http://www.plasmon.jp/reports/okamoto.pdf). That is, in devices which utilize local plasmon resonance, it is critical for the metal particles to be situated close to one another. For example, it is important for the metal particles to be mutually adjacent, without touching, at intervals therebetween of 200 nm or less.

SUMMARY OF THE INVENTION

We have conducted extensive investigations on devices which utilize local plasmon resonance. As a result, we have discovered that in prior-art devices which employ a self-ordered anodized layer, the resonance intensity is not sufficiently large.
Moreover, in the above-mentioned article by Okamoto, gold colloid particles are chemically immobilized on a glass substrate using 3-aminopropyltrimethoxysilane. However, we have found that when the resulting plasmon resonance device is used as a sample holder for Raman spectroscopy, signal strength reproducibility is poor.
It is therefore an object of the present invention to provide a structure which has a sufficiently large signal strength and generates local plasmon resonance of excellent reproducibility.
We have also found from additional studies that when a plasmon resonance device is used as a sample holder for Raman spectroscopic analysis, if a liquid sample to be analyzed does not adhere to the surface of the metal particles or adheres but does so non-uniformly, the signal strength becomes small or the signal reproducibility worsens.
We have discovered as well that, in devices which use a self-ordered anodized layer, by removing at least a portion of the anodized layer and conferring the layer with a predetermined average surface roughness (R_a), the irregularities in the filling of micropores with metal that occur during sealing treatment can be eliminated. As a result, the signal strength can be made sufficiently large, enabling an excellent reproducibility to be achieved.
Accordingly, the invention provides the following structures and the following methods for manufacturing such structures.

(1) A process for manufacturing a structure at least a portion of which is an aluminum member having on a surface thereof an anodized layer having a plurality of micropores, the process comprising the steps of, in order:
- (a) anodizing a surface of an aluminum member so as to form an anodized layer having micropores present therein;
- (b) sealing the micropores in the anodized layer by filling the micropores with metal;
- (c) surface treating the sealed anodized layer so as to remove at least a portion of upper layer surface thereof and set the average surface roughness (R_a) to at most 30 nm; and
- (d) subjecting the surface-treated anodized layer to electrodeposition so as to form metal particles on the metal filled into the micropores during sealing.
(2) The process of (1) above, wherein 20 to 80% of the anodized layer thickness is removed in the surface treatment step (c).
(3) The process of (1) above, wherein the surface treatment step (c) is carried out by at least one method selected from the group consisting of mechanical polishing, chemical dissolution, and ion beam delayering in a vacuum.
(4) The process of (3) above, wherein the mechanical polishing method is chemical mechanical polishing (CMP).
(5) A structure obtained by the process of any one of (1) to (4) above, wherein the micropores present in the anodized layer have an average diameter of 10 to 500 nm and a coefficient of variation in diameter of 5 to 20%, and the metal particles formed by electrodeposition have an average diameter larger than the average pore diameter yet smaller than an average center-to-center spacing between neighboring micropores (also referred to below as the “pore period”).

When the structure of the invention is used as a sample holder for Raman spectroscopy, because the metal particles are in close proximity and uniformly present on the structure, the local plasmon resonance is larger, greatly increasing the sensitivity and providing an excellent reproducibility.
The structure of the invention can be advantageously used also in other devices which utilize plasmon resonance.

BRIEF DESCRIPTION OF THE DIAGRAMS

FIG. 1 shows schematic sectional views of the surface of an anodized layer having a plurality of micropores following formation in an anodizing step and after being subjected to, in order, a sealing step, a surface treatment step, and an electrodeposition step.
FIG. 2 is a schematic sectional view of the surface of the inventive structure obtained when sealing is administered by electrodeposition.
FIG. 3 is a schematic sectional view of the surface of the inventive structure obtained when sealing is administered by a method which uses metal colloidal particles.
FIG. 4 is a schematic sectional view of the surface of the inventive structure shown in FIG. 2 when used as a sample holder for Raman spectroscopy.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

The inventive process for manufacturing a structure (sometimes referred to below simply as “the inventive manufacturing process”) is a process for manufacturing a structure at least a portion of which is an aluminum member having on a surface thereof an anodized layer having a plurality of micropores, the process includes the steps of, in order: (a) anodizing a surface of an aluminum member so as to form an anodized layer having micropores present therein; (b) sealing the micropores in the anodized layer by filling the micropores with metal; (c) surface treating the sealed anodized layer so as to remove at least a portion of upper layer thereof and set the average surface roughness (R_a) to at most 30 nm; and (d) subjecting the surface-treated anodized layer to electrodeposition so as to form metal particles on the metal filled into the micropores during sealing.
The structure of the invention is a structure which is obtained by the foregoing manufacturing process and in which the micropores present in the anodized layer have an average diameter of 10 to 500 nm and a coefficient of variation in diameter of 5 to 20%, and the metal particles formed by electrodeposition have an average diameter larger than the average pore diameter yet smaller than the pore period as defined above.
The aluminum member used in the invention, each of above steps (a) to (d) in the inventive manufacturing process, and the microstructure thus obtained are described in detail below.
[Aluminum Member]
The aluminum member having an anodized layer at the surface which is used in the invention can be obtained by anodizing the surface of a member having an aluminum surface.
The member having an aluminum surface is preferably at least 99.9 wt % high-purity aluminum in the portion to be anodized or the entire member. Because the member is not limited to a homogeneous material, use can also be made of a laminated material. Specific examples include aluminum substrates such as substrates made of low-purity aluminum (e.g., recycled material) on which high-purity aluminum has been vapor-deposited; substrates made of silicone wafers, quartz, glass or the like, the surface of which has been covered with high-purity aluminum by a technique such as sputtering, vapor deposition, chemical vapor deposition, electrodeposition, chemical plating or electroplating; and substrates laminated with aluminum foil.
A substrate laminated with aluminum foil can be obtained by using an intervening adhesive layer including an adhesive to apply aluminum foil onto a substrate made of resin or some other suitable material.
Specific examples of adhesives that may be used for this purpose include aromatic polyether-type one-part moisture curing adhesives (e.g., SF102RA, available from Dainippon Ink & Chemicals, Inc.); aromatic polyether-type two-part curing adhesives (e.g., 2K-SF-302A/HA550B, available from Dainippon Ink & Chemicals, Inc.); aliphatic polyester two-part curing adhesives (e.g., 2K-SF-250A/HA280B, available from Dainippon Ink & Chemicals, Inc.); water-based adhesives for dry laminates (e.g., WS305A/LB-60, WS201A/LB-60, WS325A/LJ-55, WS350A/LA-100 and WS-320A, all products of Dainippon Ink & Chemicals, Inc.); organic solvent-based adhesives for dry laminates (e.g., LX-747A/KX-75, LX-88H(T)/KW-75 and LX-732/KRX-90, all products of Dainippon Ink & Chemicals, Inc.); epoxy-based one-part heat-curable adhesives (e.g., EP106, EP138, EP160, EP170 and EP171, all products of Cemedine Co., Ltd.); one-part anaerobic-curing adhesives composed of, for example, acrylic oligomers (SGA) (e.g., the Y-800 series and Y-805 GH, all products of Cemedine Co., Ltd.); special silicone-modified polymer-based one-part elastomeric adhesives (e.g., Super-X, produced by Cemedine Co., Ltd.); phenolic resin composite polymer-based adhesives composed of a mixture of phenolic resin with butadiene or acrylonitrile rubber, any of various mixtures of phenolic resin with polyvinyl acetate, polyvinyl acetal, polyvinyl butyral or polyvinyl formal, or a mixture of phenolic resin with epoxy; two-part condensation reaction-type adhesives; two-part addition reaction-type adhesives, such as of an epoxy or an isocyanate; acrylic oligomer (SGA) and other two-part radical polymerization-type adhesives; polyimide, polyester, polyolefin and other hot-melt adhesives; rubber, polyacrylate and other pressure-sensitive adhesives; one-part normal temperature-curing adhesives composed primarily of 2-cyanoacrylates; methyl 2-cyanoacrylate adhesives; ethyl 2-cyanoacrylate adhesives (e.g., Aronalpha, produced by Toagosei Co., Ltd.); and α-cyanoacrylate adhesives (e.g., the 3000DX series available from Cemedine Co., Ltd.).
The adhesive layer has a thickness of preferably 3 to 50 μm, more preferably 5 to 20 μm, and even more preferably 10 to 20 μm. The thickness of the adhesive layer can be determined by examination of the fracture plane with a scanning electron microscope (SEM).
Aluminum foil is provided on the adhesive layer. The aluminum foil has a thickness of preferably 1 to 10 μm, more preferably 1 to 5 μm, and even more preferably 2 to 4 μm.
As will be described subsequently, when depressions intended to serve as the starting points of the main anodizing treatment are formed by a self-ordering process, a certain degree of depth is required in the member itself which has an aluminum surface, and so the use of an aluminum substrate is preferred.
Of the aluminum member, the surface on which an anodized layer is provided by anodizing treatment has an aluminum purity of preferably at least 99.5 wt %, and more preferably at least 99.80 wt %, but preferably not more than 99.99 wt %, and more preferably not more than 99.95 wt %. At an aluminum purity of at least 99.5 wt %, the pore arrangement will have sufficient regularity. At an aluminum purity of not more than 99.99 wt %, inexpensive production is possible.
The surface of the aluminum member is preferably subjected to degreasing treatment and mirror-like finishing treatment prior to anodizing treatment.
<Degreasing>
Degreasing is carried out with a suitable substance such as an acid, alkali or organic solvent so as to dissolve and remove organic substances, including dust, grease and resins, adhering to the aluminum surface, and thereby prevent defects caused by organic substances from arising in each of the subsequent treatments.
Known degreasers can be used in degreasing treatment. For example, degreasing can be carried out using any of various commercially available degreasers by the prescribed method.
Preferred methods include the following: a method in which an organic solvent such as an alcohol, a ketone, ligroin or a volatile oil is contacted with the aluminum surface at ambient temperature (organic solvent method); a method in which a liquid containing a surfactant such as soap or a neutral detergent is contacted with the aluminum surface at a temperature of from ambient temperature to 80° C., after which the surface is rinsed with water (surfactant method); a method in which an aqueous sulfuric acid solution having a concentration of 10 to 200 g/L is contacted with the aluminum surface at a temperature of from ambient temperature to 70° C. for a period of 30 to 80 seconds, following which the surface is rinsed with water; a method in which an aqueous solution of sodium hydroxide having a concentration of 5 to 20 g/L is contacted with the aluminum surface at ambient temperature for about 30 seconds while electrolysis is carried out by passing a direct current through the aluminum surface as the cathode at a current density of 1 to 10 A/dm², following which the surface is contacted with an aqueous solution of nitric acid having a concentration of 100 to 500 g/L and thereby neutralized; a method in which the aluminum surface is contacted with any of various known anodizing electrolyte solutions at ambient temperature while electrolysis is carried out by passing a direct current or an alternating current at a current density of 1 to 10 A/dm²through the aluminum surface as the cathode; a method in which an aqueous alkali solution having a concentration of 10 to 200 g/L is contacted with the aluminum surface at 40 to 50° C. for 15 to 60 seconds, following which the surface is contacted with an aqueous nitric acid solution having a concentration of 100 to 500 g/L and thereby neutralized; a method in which an emulsion prepared by mixing a surfactant, water and the like into an oil such as gas oil or kerosene is contacted with the aluminum surface at a temperature of from ambient temperature to 50° C., following which the surface is rinsed with water (emulsion degreasing method); and a method in which a mixed solution of, for example, sodium carbonate, phosphates and surfactant is contacted with the aluminum surface at a temperature of ambient temperature to 50° C. for 30 to 180 seconds, following which the surface is rinsed with water (phosphate method).
The method used for degreasing is preferably one which can remove grease from the aluminum surface but causes substantially no aluminum dissolution. Hence, an organic solvent method, surfactant method, emulsion degreasing method or phosphate method is preferred.
<Mirror-Like Finishing>
Mirror-like finishing is carried out to eliminate surface features on the aluminum member and improve the uniformity and reproducibility of sealing treatment by a means such as electrodeposition. Examples of surface features on the aluminum member include rolling streaks that form during rolling when the aluminum member has been produced by a process that includes rolling.
In the practice of the invention, mirror-like finishing is not subject to any particular limitation, and can be carried out using any suitable method known in the art. Examples of suitable methods include mechanical polishing, chemical polishing, and electrolytic polishing.
Illustrative examples of suitable mechanical polishing methods include polishing with various commercial abrasive cloths, and methods that combine the use of various commercial abrasives (e.g., diamond, alumina) with buffing. More specifically, preferred methods involving the use of abrasives include a method that is carried out while varying over time the abrasive used from coarse particles to fine particles. In such a case, the final abrasive used is preferably one having a grit size of 1500. In this way, a glossiness of at least 50% (in the case of rolled aluminum, at least 50% in both the rolling direction and the transverse direction) can be achieved.
Examples of chemical polishing methods include the various methods mentioned in the 6^thedition of Aluminum Handbook (Japan Aluminum Association, 2001), pp. 164-165.
Preferred examples include the phosphoric acid/nitric acid method, Alupol I, Alupol V, Alcoa R5, the H₃PO₄—CH₃COOH—Cu method and the H₃PO₄—HNO₃—CH₃COOH method. Of these, the phosphoric acid/nitric acid method, the H₃PO₄—CH₃COOH—Cu method and the H₃PO₄—HNO₃—CH₃COOH method are especially preferred.
With chemical polishing, a glossiness of at least 70% (in the case of rolled aluminum, at least 70% in both the rolling direction and the transverse direction) can be achieved.
Examples of electrolytic polishing methods include the various methods mentioned in the 6^thedition of Aluminum Handbook (Japan Aluminum Association, 2001), pp. 164-165.
A preferred example is the method described in U.S. Pat. No. 2,708,655.
The method described in Jitsumu Hyomen Gijutsu Vol. 33, No. 3, 32-38 (1986) is also preferred.
With electrolytic polishing, a glossiness of at least 70% (in the case of rolled aluminum, at least 70% in both the rolling direction and the transverse direction) can be achieved.
These methods can be suitably combined when used. In a preferred example, a method that uses an abrasive is carried out by changing over time the abrasive used from coarse particles to fine particles, following which electrolytic polishing is administered.
Mirror-like finishing enables a surface having, for example, an average surface roughness R_aof 0.1 μm or less and a glossiness of at least 50% to be obtained. The average surface roughness R_ais preferably 0.03 μm or less, and more preferably 0.02 μm or less. The glossiness is preferably at least 70%, and more preferably at least 80%.
The glossiness is the specular reflectance which can be determined in accordance with JIS Z8741-1997 (Method 3: 60° Specular Gloss) in a direction perpendicular to the rolling direction. Specifically, measurement is carried out using a variable angle gloss meter (e.g., VG-1D, manufactured by Nippon Denshoku Industries Co., Ltd.) at an angle of incidence/reflection angle of 60° when the specular reflectance is 70% or less, and at an angle of incidence/reflection of 20° when the specular reflectance is more than 70%.
[Anodizing Treatment Step]
The anodizing treatment step in the inventive manufacturing process is a step in which the surface of the above-described aluminum member is anodized so as to form an anodized layer having micropores present therein.
In the practice of the invention, the method of administering anodizing treatment to the surface of the aluminum member is preferably one in which, prior to the micropore-forming anodizing treatment (also referred to below as the “main anodizing treatment”), depressions are formed as starting points for micropore formation during main anodizing treatment.
<Formation of Depressions>
No particular limitation is imposed on the method of forming depressions. Examples of suitable methods include a self-ordering method which employs the self-ordering properties of the anodized layer, a physical method, a particle beam method, a block copolymer method, and a resist interference exposure method.
(Self-Ordering Method)
The self-ordering method is a method which enhances the orderliness by using the regularly arranging nature of micropores in the anodized layer and eliminating factors that disturb an orderly arrangement. Specifically, an anodized layer is formed on high-purity aluminum at a voltage appropriate for the type of electrolyte solution and at a slow speed over an extended period of time (e.g., from several hours to well over ten hours), following which delayering treatment is carried out.
In this method, because the pore diameter is dependent on the voltage, the desired pore diameter can be obtained to a certain degree by controlling the voltage.
Typical examples of self-ordering methods include those described in J. Electrochem. Soc. Vol. 14, No. 5, L128 (May 1997); Jpn. J. Appl. Phys. Vol. 35, Part 2, No. 1B, L126 (1996); Appl. Phys. Lett. Vol. 71, No. 19, 2771 (Nov. 10, 1997), and in the above-referenced article by Masuda (1998). These self-ordering methods are carried out under the conditions indicated below.

(1) 0.3 mol/L sulfuric acid, 0° C., 27 V, 450 minutes (J. Electrochem. Soc., 1997)
(2) 0.3 mol/L sulfuric acid, 10° C., 25 V, 750 minutes (J. Electrochem. Soc., 1997)
(3) 0.3 mol/L oxalic acid, 17° C., 40 V, 600 minutes; followed by pore widening treatment (solution containing 6 wt % phosphoric acid and 1.8 wt % chromic acid, 60° C., 840 minutes) (Jpn. J. Appl. Phys., 1996)
(4) 0.3 mol/L oxalic acid, 17° C., 40 to 60 V, 36 minutes; followed by pore widening treatment (5 wt % phosphoric acid, 30° C., 70 minutes) (Appl. Phys. Lett., 1997)
(5) 0.04 mol/L oxalic acid, 3° C., 80 V, layer thickness 3 μm; followed by pore widening treatment (5 wt % phosphoric acid, 30° C., 70 minutes) (Appl. Phys. Lett., 1997)
(6) 0.3 mol/L phosphoric acid, 0° C., 195 V, 960 minutes; followed by pore widening treatment (10 wt % phosphoric acid, 240 minutes) (Masuda, 1998).

In the methods described in these prior-art publications, delayering treatment to dissolve and remove the anodized layer is applied for at least 12 hours using a mixed aqueous solution of chromic acid and phosphoric acid at about 50° C. Carrying out treatment using a boiling water solution destroys or disrupts the starting points for self-ordering. Hence, the aqueous solution is used without being boiled.
The orderliness of the self-ordered anodized layer increases as the underlying aluminum is approached; once delayering has taken place, the lower portion of the anodized layer remaining on the underlying aluminum emerges at the surface, affording an orderly arrangement of depressions. Therefore, in delayering treatment, only the anodized layer made of aluminum oxide is dissolved; the aluminum is not dissolved.
The self-ordering anodizing treatment used in this invention may be carried out by, for example, a method that involves passing an electrical current through the aluminum member as the anode within a solution having an acid concentration of 1 to 10 wt %. Solutions that may used in anodizing treatment include any one or combinations of two or more of the following: sulfuric acid, phosphoric acid, chromic acid, oxalic acid, sulfamic acid, benzenesulfonic acid and amidosulfonic acid.
The conditions of the self-ordering anodizing treatment vary empirically with the electrolyte solution used, although it is generally suitable for the electrolyte concentration to be 1 to 10 wt %, the temperature of the solution to be 0 to 20° C., the current density to be 0.1 to 10 A/dm², the voltage to be 10 to 200 V, and the period of electrolysis to be 2 to 20 hours.
The self-ordered anodized layer has a thickness of preferably 10 to 50 μm.
In the practice of the invention, the self-ordering anodizing treatment is carried out for a period of preferably 1 to 16 hours, more preferably 2 to 12 hours, and even more preferably 2 to 7 hours.
Delayering treatment is carried out for a period of preferably 0.5 to 10 hours, more preferably 2 to 10 hours, and even more preferably 4 to 10 hours.
After an anodized layer has been formed in this way by a self-ordering method, if the layer is then dissolved and removed, and the subsequently described main anodizing treatment is carried out under the same conditions, substantially straight micropores will form substantially perpendicular to the surface of the layer.
(Physical Method)
Physical methods are exemplified by methods which use press patterning. A specific example is a method in which a substrate having on the surface a plurality of protrusions is pressed against the aluminum surface to form depressions therein. For instance, the method described in JP 10-121292 A can be used.
Another example is a method in which polystyrene spheres are densely arranged on the aluminum surface, SiO₂is vapor-deposited over the layer of the spheres, then the polystyrene spheres are removed and the substrate is etched using the vapor-deposited SiO₂as the mask, thereby forming depressions.
(Particle Beam Method)
In the particle beam method, depressions are formed by irradiating the aluminum surface with a particle beam. This method has the advantage that the positions of the depressions can be freely controlled.
Examples of the particle beam include a charged particle beam, a focused ion beam (FIB), and an electron beam.
An example of a particle beam method that can be used is the method described in JP 2001-105400 A.
(Block Copolymer Method)
The block copolymer method is a method that involves forming a block copolymer layer on the aluminum surface, forming an islands-in-the-sea structure in the block copolymer layer by thermal annealing, then removing the island components to form depressions.
An example of a block copolymer method that can be used is the method described in JP 2003-129288 A.
(Resist Interference Exposure Method)
In the resist interference exposure method, a resist is provided on the aluminum surface, then the resist is exposed and developed so as to form depressions which pass entirely through the resist to the aluminum surface.
An example of a resist interference exposure method that can be used is the method described in JP 2000-315785 A.
Of the various above methods of forming depressions, the self-ordering method, FIB method and resist interference exposure method are desirable because they are capable of uniformly forming depressions over a large surface area of about 10 cm square or more.
From the standpoint of production costs, the self-ordering method is especially preferred. The FIB method is also desirable because it enables the arrangement of micropores to be controlled at will.
The depressions formed have a depth of preferably at least about 10 nm and a width which is preferably not greater than a width of the desired pore diameter.
<Main Anodizing Treatment>
In the practice of the invention, after depressions have been formed as desired in the aluminum surface, an anodized layer having micropores therein is formed by main anodizing treatment.
A method known in the art can be used to carry out this anodizing treatment.
Specifically, suitable use can be made of a method in which the current is repeatedly turned on and off while maintaining a constant dc voltage, or a method in which the current is repeatedly turned on and off while intermittently varying the dc voltage. These methods are preferred because they form fine micropores in the anodized layer, thus enhancing the uniformity, particularly during sealing treatment by electrodeposition.
In the above-described method in which the voltage is intermittently varied, it is preferable to successively lower the voltage. By doing so, the resistance of the anodized layer can be lowered, enabling uniformity to be achieved when electrodeposition is subsequently carried out.
When this main anodizing treatment is carried out at a low temperature, the arrangement of micropores is orderly made and the pore diameter is uniform.
In the view point of easiness of filling the micropores with metal, it is preferable for the anodized layer to have a thickness which is from 0.5 to 10 times, preferably 1 to 8 times, and even more preferably 1 to 5 times, the pore diameter.
If electrodeposition is to be subsequently carried out as the sealing treatment, it is preferable for the pore diameter to be at least 10 nm.
Therefore, according to a preferred embodiment, the anodized layer has a thickness of 0.1 to 1 μm and the micropores have an average diameter of 10 to 500 nm.
It is preferable for the micropores to have a coefficient of variation in pore diameter of 5 to 20%. At a coefficient of variation in pore diameter within this range, the filling efficiency in the subsequently described sealing step will be high and the metal particles formed thereafter will be present in close proximity each other on the structure ultimately obtained, increasing the local plasmon resonance.
Here, the coefficient of variation in pore diameter is an indicator of the variation in the pore diameter, and is defined as follows.
Pore-diameter coefficient of variation in %=(pore diameter standard deviation)/(average pore diameter)×100
The micropores have an average pore density of preferably 50 to 1,500 pores/μm².
The micropores have a surface coverage of preferably 20 to 50%. The surface coverage of the micropores is defined as the ratio of the total surface area of the micropore openings with respect to the surface area of the aluminum surface. When computing the surface coverage by the micropores, all micropores are included, whether or not they are filled with metal. This value is determined by measuring the surface porosity prior to the sealing step.
<Pore-Widening Treatment>
In the practice of the invention, following the main anodizing treatment, when necessary, it is desirable to carry out pore-widening treatment by immersing the aluminum member in an aqueous solution of an acid or an alkali so as to dissolve the anodized layer and enlarge the diameter of the micropores.
It is possible in this way to dissolve the barrier film at the bottom of the micropores in the anodized layer and thereby selectively effect electrodeposition within the micropores.
When pore-widening treatment is to be carried out with an aqueous acid solution, it is preferable to use an aqueous solution of an inorganic acid such as sulfuric acid, phosphoric acid, nitric acid or hydrochloric acid, or a mixture thereof. It is desirable for the aqueous acid solution to have a concentration of 1 to 10 wt % and a temperature of 25 to 40° C.
When pore widening treatment is to be carried out with an aqueous alkali solution, it is preferable to use an aqueous solution of at least one alkali selected from the group consisting of sodium hydroxide, potassium hydroxide and lithium hydroxide. It is desirable for the aqueous alkali solution to have a concentration of preferably 0.1 to 5 wt % and a temperature of preferably 20 to 35° C.
Specific examples of preferred solutions include a 40° C. aqueous solution containing 50 g/L of phosphoric acid, a 30° C. aqueous solution containing 0.5 g/L of sodium hydroxide, and a 30° C. aqueous solution containing 0.5 g/L of potassium hydroxide.
The immersion time within the aqueous solution of an acid or an alkali is preferably 8 to 60 minutes, more preferably 10 to 50 minutes, and even more preferably 15 to 30 minutes.
<Other Treatment>
If necessary, other treatment may also be administered.
For example, if the inventive microstructure is to be used as a sample holder on which a drop of an aqueous solution will be placed and formed into a film, hydrophilizing treatment may also be administered to lower the contact angle with water. Such hydrophilizing treatment may be carried out by a method known in the art.
Alternatively, if the inventive microstructure is to be used as a sample holder for proteins which are denatured or decomposed by acids, neutralizing treatment may be administered to neutralize any residues of the acid used in the main anodizing treatment that remain thereafter on the aluminum surface. Such neutralizing treatment can be carried out by a method known in the art.
[Sealing Step]
The sealing step of the inventive method of manufacture is a step in which the micropores are filled with metal following the above-described anodizing treatment step.
In the practice of the invention, the metal is not subject to any particular limitation, so long as it is an element which has metal bonds with free electrons and which is electrically conductive. However, a metal in which plasmon resonance has been recognized is preferred.
Specific examples include copper, nickel, chromium, zinc, gold, silver, platinum and palladium (see Mekki Ky{overscore (o)}hon [Plating manual] (Nikkan Kogyo Shimbun Co., Ltd., 2001), pp. 63-153.
Of these, copper, nickel and gold are preferred on account of their high electrical conductivity and high uniformity. Nickel and gold are more preferred because of their adherence with the metal particles to be provided later.
Moreover, it is known that copper, nickel, gold, silver and platinum readily give rise to plasmon resonance (Gendai Kagaku, pp. 20-27 (September 2003)). Gold and silver are especially preferred because of the ease of electrodeposition and colloidal particle formation.
Sealing may be carried out using any suitable known technique without particular limitation. Examples of preferred techniques include electrodeposition, methods which involve coating the aluminum member having an anodized layer with a dispersion of metal colloidal particles, then drying; and electroless plating. The metal is preferably in the form of single particles or agglomerates.
<Electrodeposition>
An electrodeposition method already known in the art may be used. Specific examples of gold electrodeposition methods include a method in which the anodized aluminum member is immersed in a 30° C. dispersion containing 1 g/L of HAuCl₄and 7 g/L of H₂SO₄and electrodeposition is carried out at a constant voltage of 11 V (adjusted with a variable transformer) for 5 to 6 minutes; and a method in which the anodized aluminum member is immersed in a dispersion (liquid temperature, 40° C.; pH, 6) containing 6 g/L of potassium dicyanoaurate (I) and 10 g/L of a chelating agent (EDTA), and electrodeposition is carried out for 5 to 6 minutes at a current density of 0.1 A/dm²using a high-purity platinum plate as the counterelectrode.
An example of an electrodeposition method which uses copper, tin and nickel is described in detail in Gendai Kagaku, pp. 51-54 (January 1997)). Use can be made of this method as well.
<Sealing Method which Use Metal Colloidal Particles>
The dispersions employed in sealing methods which use metal colloidal particles can be obtained by a known method. Illustrative examples include methods of preparing fine particles by low-vacuum vapor deposition and methods of preparing metal colloids by reducing an aqueous solution of a metal salt.
The metal colloidal particles have an average particle size of preferably 1 to 200 nm, more preferably 1 to 100 nm, and even more preferably 2 to 80 nm.
Preferred use can be made of water as the dispersion medium employed in the dispersion. Use can also be made of a mixed solvent composed of water and a solvent that is miscible with water, such as an alcohol, illustrative examples of which include ethyl alcohol, n-propyl alcohol, i-propyl alcohol, 1-butyl alcohol, 2-butyl alcohol, t-butyl alcohol, methyl cellosolve and butyl cellosolve.
No particular limitation is imposed on the technique used for coating the aluminum member having an anodized layer with the dispersion of metal colloidal particles. Suitable examples of such techniques include bar coating, spin coating, spray coating, curtain coating, dip coating, air knife coating, blade coating and roll coating.
Preferred examples of dispersions that may be employed in sealing methods which use metal colloidal particles include dispersions of gold colloidal particles and dispersions of silver colloidal particles.
Gold colloidal particle dispersions that may be used include those described in JP 2001-89140 A and JP 11-80647 A. Use can also be made of commercial products.
Dispersions of silver colloidal particles preferably contain particles of silver-palladium alloys because these are not affected by the acids which dissolve out from the anodized layer. The palladium content in such a case is preferably from 5 to 30 wt %.
After the anodized aluminum member has been coated with the dispersion, it may be suitably cleaned using a solvent such as water. As a result of such cleaning, only the particles filled into the micropores remain on the anodized layer; particles that have not been filled into micropores are removed.
<Electroless Plating>
An electroless plating method already known in the art may be used. Electroless plating can be used if the micropores on the anodized layer have a diameter of at least 300 nm. Examples of preferred electroless plating methods include those which use a gold cyanide complex and a reducing agent.
The mechanism of this electroless plating method, while not well understood, appears to proceed as follows. Because the metal and the reducing agent readily react, reduction of the barrier film occurs, leaving the underlying aluminum metal exposed. Due to a substitution reaction between the underlying aluminum metal and the gold cyanide complex, some of the aluminum is substituted with gold so that the underlying aluminum metal surface becomes coated with gold. Once the surface is covered with gold, the gold grows due to an autocatalyzed reaction, enabling the micropores in the anodized layer to be filled with gold without having to resort to electrolysis. The anodized layer obtained after treatment thus has a high electrical conductivity and is uniform.
In the practice of the invention, the amount of metal deposited after sealing is preferably 100 to 500 mg/m².
The surface porosity following sealing treatment is preferably not more than 20%. The surface porosity following sealing treatment is defined as the total surface area of the openings in unfilled micropores with respect to the surface area of the aluminum surface. When the surface porosity is in the foregoing range, a stronger local plasmon resonance is obtained.
At a pore diameter of 50 nm or more, it is preferable to use a sealing method that employs metal colloidal particles. At a pore diameter of less than 50 nm, the use of an electrodeposition process is preferred. Suitable use can also be made of a combination of both.
[Surface Treatment Step]
The surface treatment step in the inventive method of manufacture is a step which, following the above-described sealing step, removes at least a portion of upper layer surface of the anodized layer and sets the average surface roughness (R_a) to at most 30 nm.
In the practice of the invention, no particular limitation is imposed on removal of the portion of upper layer surface of the anodized layer, provided such removal is carried out so that the average surface roughness (R_a) becomes at most 30 nm. However, removal of 20 to 80% of the thickness of the anodized layer prior to removal is preferred. The average surface roughness (R_a) is preferably at least 5 nm, and more preferably from 10 to 20 nm.
Also, in practicing the invention, removal of the portion of upper layer surface of the anodized layer is preferably carried out by a least one method selected from the group consisting of the subsequently described mechanical polishing, chemical dissolution and ion beam delayering in a vacuum. Removal by mechanical polishing, and especially chemical mechanical polishing (CMP), is more preferred.
<Mechanical Polishing>
Mechanical polishing is not subject to any particular limitation. Use can be made of the method employed in the above-described mirror-like finishing treatment. Of these, chemical mechanical polishing (CMP) is preferred.
CMP processes that may be used include known CMP polishing techniques commonly employed for polishing semiconductor silicon wafers.
Here, the CMP polishing technique is characterized in that, compared with prior-art polishing techniques which use general-purpose polishing platens, the platen itself has a high precision, the sample holding method and stability are excellent, the abrasive has a small particle size, and the liquid used during polishing dissolves the anodized layer on the sample.
Specifically, it is difficult with a general-purpose platen to control the planarity, the holding of the sample or the like in a nanometer order. Moreover, when an ordinary abrasive is used on a general-purpose platen, the particles are too coarse (average particle size, 0.06 to 10 μm) and tend to destroy micropores present in the anodized layer, making it difficult to achieve nanometer-order planarization. By contrast, with a CMP platen on which ultrafine particles having a submicron-order particle diameter (average particle diameter, 0.01 to 1 μm) can be used as the abrasive, the portion of upper layer of the anodized layer can be easily removed without destroying the structure of the anodized layer.
Also, in the CMP method, the micropores present in the anodized layer have a wall thickness, when sufficiently filled with metal, of preferably 10 to 100 nm.
When polishing is carried out by the CMP technique, an acidic solution or an alkaline solution may be furnished to the platen. Examples of acidic solutions that may be used include phosphoric acid, sulfuric acid, nitric acid and hydrochloric acid. Examples of alkaline solutions that may be used include potassium hydroxide and ammonium hydroxide.
Abrasives that may be used include any that are ordinarily used in CMP methods without particular limitation. Examples include diamond, SiC, Al₂O₃, SiO₂and ZrO₂.
Detailed descriptions of examples of CMP techniques, including the tools and abrasives used, appear in CMP no Saiensu [The science of CMP] (Science Forum, 1999), pp. 72-98, 123-176). Use can be made of the methods described there.
<Chemical Dissolution (Chemical Polishing)>
“Chemical dissolution method,” as used herein, refers to a method in which the anodized layer is dissolved using a chemical (solution) having the ability to dissolve aluminum oxide. Specific examples of such chemical (solution)s include sulfuric acid, chromic acid, oxalic acid, phosphoric acid, and combinations of chromic acid and phosphoric acid. These may be used singly or as combinations of two or more thereof.
Dissolution by a chemical dissolution method proceeds by the selective removal of places not completely filled with metal and enlargement of the micropores. Hence, the thickness of the anodized layer itself does not substantially change. The dissolution rate is generally about 10 nm/min to 100 nm/min.
In a chemical dissolution method, the micropores present in the anodized layer have an average thickness of partition wall of the two adjacent pores of preferably 50 to 500 nm.
Use can also be made of any techniques that may be employed in the above-described mirror-like finishing treatment.
<Ion Beam Delayering in a Vacuum>
“Ion beam delayering in a vacuum” refers herein to a method for the atom-order removal of the target substance (the portion of upper-layer surface of the above-described anodized layer) by irradiation with ions in a vacuum. This method enables a very uniform and clean surface to be obtained. However, the apparatus is generally expensive. Moreover, this technique cannot be used to treat a large surface area, and so the production efficiency is sometimes poor.
Examples of ion beam delayering in a vacuum are described in considerable detail in Jitsuy{overscore (o)} Seimitsu-kak{overscore (o)} to Keisoku Gijitsu [Practical precision machining and instrumentation techniques] (NTS, 2003), pp. 193-210. Use can also be made of these methods.
The thickness of the anodized layer after such mechanical polishing, chemical dissolution, or ion beam delayering in a vacuum can be measured with a scanning electron microscope (SEM), and the average surface roughness (R_a) can be measured using an atomic force microscope (AFM).
Specifically, as shown in the Examples described later in the specification, the thickness of the anodized layer is measured by cutting away portions of the sample before and after layer removal, bending each portion 90 degrees and examining the fracture plane where the specimen has cracked with a field-effect scanning electron microscope (FE-SEM) at a suitable magnification. R_aof the anodized layer following surface treatment is measured using AFM with a cantilever.
[Electrodeposition Step]
The electrodeposition step in the inventive method of manufacture is a step in which, following the surface treatment step described above, additional metal particles are formed by electrodeposition on the metal that has been filled into the micropores by the above-described sealing treatment.
In the inventive method, the metal particles are exemplified by particles of the metals mentioned above in the sealing step. Gold particles, silver particles and nickel particles are preferred. Gold particles are especially preferred because of their high electrical conductivity and resistance to oxidation.
As with the electrodeposition methods mentioned above in the description of the sealing step, any suitable method known in the art may be used as the electrodeposition method used in the electrodeposition step. In particular, when gold particles are to be formed, examples of methods that can be used include a method in which, as described above, the aluminum member is immersed in a 30° C. dispersion containing 1 g/L of HAuCl₄and 7 g/L of H₂SO₄, and electrodeposition treatment is carried out for 5 to 6 minutes at a constant voltage of 11 V (adjusted with a variable transformer); and a method in which the aluminum member is immersed in a dispersion (liquid temperature, 40° C.; pH, 6) containing 6 g/L of potassium dicyanoaurate (I) and 10 g/L of a chelating agent (EDTA), and electrodeposition is carried out for 5 to 6 minutes at a current density of 0.1 A/dm²using a high-purity platinum plate as the counterelectrode. Similar effects can also be obtained using a method that involves the electrodeposition of nickel (nickel electrodeposition).
Here, a known nickel electrodeposition method like any of those described in the above-referenced Mekki Ky{overscore (o)}hon [Plating manual] (Nikkan Kogyo Shimbun Co., Ltd., 2001), pp. 81-89 can be used for nickel electrodeposition.
Specifically, it is preferable to use as the nickel ion source (generally containing cobalt in a ratio of about 1/100), for example: nickel sulfate in an amount of 50 to 400 g/L, preferably 150 to 390 g/L, and more preferably 220 to 380 g/L; and nickel chloride in an amount of 10 to 100 g/L, preferably 20 to 80 g/L, and more preferably 30 to 60 g/L. In addition, boric acid capable of adjusting the pH to from 3 to 6 can be used as a pH regulator in an amount of 15 to 45 g/L, preferably 20 to 45 g/L, and more preferably 30 to 40 g/L. Also, the electrolysis conditions are preferably a temperature of 25 to 65° C., a current density of 0.5 to 10 A/dm², and a voltage of 20 to 120 V. The anode electrode material may be, for example, high-purity nickel (≧99.99%), high-purity platinum (≧99.9%), or titanium plated with platinum (≧99.9%). Of these, high-purity nickel is especially preferred.
In the inventive structure, the particles which are electrodeposited in such a electrodeposition step have an average particle size which is larger than the average pore diameter of the micropores present in the anodized layer, yet smaller than the pore period which is the average center-to-center spacing between neighboring micropores.
Here, the average particle size of the metal particles is determined by using a FE-SEM microscope to examine the surface of the structure following the electrodeposition step, measuring the diameters of 100 particles within a suitable region, and computing the average of those values.
The pore period can be obtained by using a FE-SEM microscope to examine the surface of the structure before the electrodeposition step, measuring the center-to-center intervals between 100 neighboring micropores in a suitable region, and computing the average of the measured values.
FIG. 1 shows schematic sectional views of the surface of the anodized layer having micropores as it appears after the above-described anodizing treatment step, and as it appears after, respectively, the above-described sealing step, the surface treatment step, and the electrodeposition step.
FIG. 1A is a schematic sectional view of an anodized layer 1 in which are present micropores 2 that were formed in the anodizing treatment step.
FIG. 1B is a schematic sectional view of an anodized layer 1 showing the micropores 2 filled with a first metal 3 in the sealing step (electrodeposition). As shown in FIG. 1B, filling by the first metal 3 in the sealing step is not carried out uniformly in all the micropores 2; some micropores are mostly filled, while others are partly filled but remain largely empty (in FIG. 1B, note the striking contrast between the micropore at the center and the micropore on the right). Such uneven filling is believed to be due to irregularities in the thickness of the insulating barrier film present at the bottom of the anodized layer 1.
FIG. 1C is a schematic sectional view of the anodized layer 1 when the portion of upper layer surface of the anodized layer 1 has been removed in the surface treatment step and the surface has been set to a given average surface roughness (R_a). In the practice of the invention, this surface treatment step has the effect of placing each micropore 2 in a state filled to the surface with the first metal 3, thus reducing the lack of uniformity in metal filling that arose in the earlier sealing step described above.
FIG. 1D is a schematic sectional view of the anodized layer 1, showing the additional formation in the electrodeposition step of a second metal 4 on the first metal 3 within the micropores 2. As shown in FIG. 1D, in the electrodeposition step which follows the surface treatment step, the second metal 4 is formed uniformly in the manner of particles which rise up from the surface of the anodized layer 1.
FIG. 1E is a schematic sectional view of the anodized layer 1 obtained when electrodeposition is administered without first carrying out a surface treatment step. In the absence of a surface treatment step, electrodeposition is administered to an anodized layer surface in which the micropores remain unevenly filled with the first metal 3 from the sealing step; the second metal 4 is formed on the first metal 3 within the micropores, but because variations arise in its shape, all of the second metal 4 does not take on a raised, particle-like shape like that shown in FIG. 1D.
[Microstructure]
The structure of the invention is obtained by a manufacturing method having the above-described anodizing step, sealing step, surface treatment step and electrodeposition step, and is a structure in which the micropores present in the anodized layer have an average diameter of 10 to 500 nm and a coefficient of variation in diameter of 5 to 20%, and the metal particles have an average diameter larger than the average pore diameter yet smaller than the pore period.
FIG. 2 is a schematic sectional view showing the surface of a structure according to the invention which has been obtained by using an electrodeposition method to administer sealing treatment. In the structure 10 shown in FIG. 2, micropores 14 in the anodized layer 12 are filled with a first metal 16. After a surface treatment step and an electrodeposition step have been carried out, the surface of a second metal 17 takes on a particle-like shape which rises up from the surface of the anodized layer 12.
FIG. 3 is a schematic sectional view showing the surface of a structure according to the invention which has been obtained by using a method that employs metal colloidal particles to administer sealing treatment. As with the above-described structure 10, in the structure 20 shown in FIG. 3, micropores 24 in an anodized layer 22 are filled with a first metal 26. Moreover, after a surface treatment step and an electrodeposition step have been carried out, the surface of a second metal 27 takes on a particle-like shape which rises up from the surface of the anodized layer 22. In some cases, voids remain at the interior of the micropores 24 (this is most striking in the micropore on the right side of the diagram in FIG. 3).
It is generally preferable for the intervals between these metal particles to be short so as to increase Raman enhancement. The optimal interval is affected by the size and shape of the metal particles.
Hence, the interval between the metal particles varies empirically, although it is generally preferable for the interval to be in a range of 1 to 400 nm, more preferably 5 to 300 nm, and even more preferably 10 to 200 nm. Within the above range, Raman enhancement increases.
As used herein, “metal particle interval” refers to the shortest distance between the surfaces of neighboring particles.
<Raman Enhancement Owing to Local Plasmon Resonance>
Raman enhancement refers to an effect in which the Raman scattering intensity of molecules adsorbed onto the metal is enhanced by a factor of about 10⁵to 10⁶, and is called “surface-enhanced Raman scattering” (SERS). The above-referenced publication Mekki Ky{overscore (o)}hon [Plating manual] (2001) states that Raman enhancement can be obtained by local plasmon resonance using particles of metals such as gold, silver, copper, platinum and nickel.
FIG. 4 is a schematic sectional view of the surface of the inventive structure shown in FIG. 2 which is intended for use as a sample holder for Raman spectroscopy.
If the Raman spectroscopy specimen is a hydrophilic substance, it is preferable for a structure 10 in which the surface of the anodized layer 12 is hydrophobic to be used. If the Raman spectroscopy specimen is a hydrophobic substance, it is preferable for a structure in which the surface of the anodized layer 12 is hydrophilic to be used. In this way, as shown in FIG. 4, the specimen 18 will selectively and uniformly adhere to the surface of the particle-like second metal 17.
When the inventive structure is used as a sample holder for Raman spectroscopy, the signal intensity of Raman scattering becomes sufficiently large and the reproducibility is excellent. This appears to be accounted for as follows. By removing at least a portion of the anodized layer, a large number of metal particles which protrude outward in the shape of raised, dome-like particles as shown in FIG. 1D are present, facilitating control of the interval between the metal particle to metal particle within an optimal range. The result is a higher signal intensity and good reproducibility.
The inventive structure is used as a sample holder for Raman spectroscopy in much the same way as prior-art sample holders for Raman spectroscopy. Specifically, by irradiating the Raman spectroscopy sample holder with light and measuring the Raman scattering intensity of the reflected or transmitted light, the properties of a substance which is held on the sample holder and is located next to the metal is detected.

EXAMPLES

Examples are given below by way of illustration and should not be construed as limiting the invention.

Examples 1 to 27, Comparative Examples 1 and 2

1. Fabrication of Structure
As shown in Table 1, substrates were subjected to, in order, mirror-like finishing, depression forming treatment, main anodizing treatment, pore widening treatment, sealing, surface treatment and electrodeposition, thereby giving various structures.

In Table 1, designations such as “2-1” and “2-2” in the Self-Ordered Anodizing Treatment column indicate the conditions shown in Table 2, designations such as “4-1” and “4-2” in the Main Anodizing Treatment column indicate the conditions shown in Table 4, designations such as “1” and “2” in the Sealing column indicate, for example, Sealing Treatment 1 and Sealing Treatment 2 described later in the specification, and “--” indicates that the treatment in question was not administered.

	TABLE 1


	Electrodeposition Conditions

	Self-					Average
	Ordered	Main		Surface Treatment		particle

Anodizing	Anodizing		Delayering	Time	Metal	size	Electrodeposition
Treatment	Treatment	Sealing	method	(s)	particles	(nm)	time (s)

Example 1	2-1	4-1	2	A	30 + 10	Au	800	1,000
Example 2	2-2	4-2	1	A	30 + 10	Au	500	700
Example 3	2-2	4-2	2	A	30 + 10	Au	500	700
Example 4	2-2	4-2	2	A	30 + 10	Au	500	700
Example 5	2-2	4-2	2	A	30 + 10	Au	500	700
Example 6	2-5	4-5	2	A	30 + 10	Au	180	220
Example 7	2-5	4-5	2	B1	900	Au	80	60
Example 8	2-5	4-5	2	B1	900	Au	40	30
Example 9	2-5	4-5	1	B1	900	Au	40	30
Example 10	2-5	4-5	2	B2	300	Au	40	30
Example 11	2-5	4-5	2	C	1,800	Au	30	20
Example 12	2-2	4-2	3	A	30 + 10	Au	500	700
Example 13	2-2	4-2	4	A	30 + 10	Au	500	700
Example 14	2-2	4-2	5	A	30 + 10	Au	500	700
Example 15	2-2	4-2	6	A	30 + 10	Au	500	700
Example 16	2-2	4-2	1	A	30 + 10	Au	500	700
Example 17	2-2	4-2	7	A	30 + 10	Au	500	700
Example 18	2-2	4-2	8	A	30 + 10	Au	500	700
Example 19	2-2	4-2	9	A	30 + 10	Au	500	700
Example 20	2-2	4-2	2	A	30 + 10	Cu	500	60
Example 21	2-2	4-2	2	A	30 + 10	Ni	500	30
Example 22	2-2	4-2	2	A	30 + 10	Cr	500	2
Example 23	2-2	4-2	2	A	30 + 10	Zn	500	60
Example 24	2-2	4-2	2	A	30 + 10	Au	500	600
Example 25	2-2	4-2	2	A	30 + 10	Ag	500	120
Example 26	2-2	4-2	2	A	30 + 10	Pt	500	3
Example 27	2-2	4-2	2	A	30 + 10	Pd	500	120
Comp. Ex. 1	2-1	4-1	2	—	—	Au	800	1,000
Comp. Ex. 2	2-2	4-2	2	—	—	Au	500	700

The substrate and various treatments are described below.
(1) Substrate
The following Substrate 1 was used to fabricate the structure in each Example.
Substrate 1: High-purity aluminum produced by Wako Pure Chemical Industries, Ltd.; purity, 99.99 wt %; thickness, 0.4 mm.
(2) Mirror-Like Finishing
Above Substrate 1 was subjected to the following mirror-like finishing treatment.
<Mirror-Like Finishing>
Mirror-like finishing was administered by carrying out polishing with an abrasive cloth, buffing, then electrolytic polishing, in this order. After buffing, the substrate was rinsed with water.
Polishing with an abrasive cloth was carried out using a polishing platen (Abramin, produced by Marumoto Struers K.K.) and commercial water-resistant abrasive cloths. This polishing operation was carried out while successively changing the grit of the water-resistant abrasive cloths in the following order: #200, #500, #800, #1000 and #1500.
Buffing was carried out using slurry-type abrasives (FM No. 3 (average particle size, 1 μm) and FM No. 4 (average particle size, 0.3 μm), both made by Fujimi Incorporated).
Electrolytic polishing was carried out for 2 minutes using an electrolyte solution of the composition indicated below (temperature, 70° C.), using the substrate as the anode and a carbon electrode as the cathode, and at a constant current of 130 mA/cm². The power supply was a GP0110-30R unit manufactured by Takasago, Ltd.
<Electrolyte Solution Composition>

85 wt % Phosphoric acid (Wako Pure Chemical Industries) 660 mL

Pure water 160 mL

Sulfuric acid 150 mL

Ethylene glycol 30 mL

(3) Formation of Depressions
The following self-ordering method was used to form, in the surface of the mirror-like finished Substrate 1, depressions intended to serve as the starting points of micropore formation in the subsequently described main anodizing treatment.
Self-Ordering Method

Self-ordering anodizing treatment was carried out under one of the sets of conditions (type, concentration and temperature of electrolyte solution, voltage, current density, and treatment time) shown in Table 2 as 2-1 to 2-10, thereby forming anodized layers of the thicknesses shown in Table 2. In self-ordered anodizing treatment, use was made of NeoCool BD36 (Yamato Scientific Co., Ltd.) as the cooling system, Pairstirrer PS-100 (made by EYELA) as the stirring and warming unit, and a GP0650 unit (Takasago, Ltd.) as the power supply.

TABLE 2


					Current	Treatment	Layer
	Type of	Concentration	Temperature	Voltage	density	time	thickness
Conditions	electrolyte	(mol/L)	(° C.)	(V)	(A/dm²)	(hours)	(μm)

2-1	phosphoric acid	0.3	0	195	0.25	8.0	40
2-2	phosphoric acid	0.3	7	130	0.60	4.0	50
2-3	phosphoric acid	1.0	7	80	0.50	4.0	40
2-4	phosphoric acid	1.0	20	16	0.10	10.0	20
2-5	oxalic acid	0.3	16	40	1.20	2.0	60
2-6	oxalic acid	0.3	25	40	2.40	1.5	40
2-7	sulfuric acid	0.3	0	27	1.20	3.5	50
2-8	sulfuric acid	0.3	10	25	2.00	6.5	130
2-9	sulfuric acid	0.3	16	25	3.00	1.0	15
2-10	sulfuric acid	1.0	25	10	4.00	0.5	10

The phosphoric acid, oxalic acid and sulfuric acid in Table 2 were all reagents made by Kanto Chemical Co., Inc. The current densities indicated are the values when the current was stable.
Next, the substrate on which an anodized layer had been formed was immersed in a treatment solution under the conditions shown in Table 3, thereby carrying out delayering treatment in which the anodized layer is dissolved.
The delayering rate was computed as follows from the change over time in the thickness of the anodized layer and from the treatment time. The results are shown in Table 3. The thickness of the anodized layer after delayering treatment was less than 0.1 μm in each case.
<Computing the Delayering Rate>
Substrates sampled at one-hour intervals during delayering treatment were bent, the fracture plane where the specimen cracked was examined with an ultrahigh-resolution SEM (Hitachi S-900, manufactured by Hitachi, Ltd.) at a relatively low acceleration voltage of 12 V and without administering, for example, vapor-deposition treatment to confer electrical conductivity, and the layer thickness was measured. Sampling involved randomly selecting ten pieces each time, and determining the average layer thickness. The layer thickness error was in a range of ±10%.

TABLE 3

85 wt % Anhydrous

Phosphoric chromic Delayering

acid acid Pure water Temp. rate

(g) (g) (g) (° C.) (μm/h)

100 30 1500 50 4
In Table 3, the 85 wt % phosphoric acid and anhydrous chromic acid were reagents produced by Kanto Chemical Co., Inc.
(4) Main Anodizing Treatment
The substrate on which depressions had been formed were then subjected to main anodizing treatment. Main anodizing treatment was carried out by immersing the substrate in an electrolyte solution, and carrying out electrolytic treatment one or more times under one of the sets of conditions (type, concentration and temperature of electrolyte solution, voltage in first electrolytic treatment) shown in Table 4 under the designations 4-1 to 4-9.
When electrolytic treatment was carried out a plurality of times, in the first treatment, electrolysis was interrupted when the voltage reached an initial voltage setting V₀; in the second treatment, electrolysis was interrupted when the voltage reached 0.9 times the initial voltage setting (0.9×V₀); in the third treatment, electrolysis was interrupted when the voltage reached 0.8 times the initial voltage setting (0.8×V₀); in the nth treatment, electrolysis was interrupted when the voltage reached {1−0.1×(n−1)} times the initial voltage setting ({1−0.1×(n−1)}×V₀).

The thickness of the anodized layer was measured in the same way as described above. The resulting increments in thickness are shown in Table 4.

TABLE 4


				Voltage in
				first	Number	Increment in film
	Type of	Concentration	Temp.	electrolysis	of	thickness
Condition	electrolyte	(mol/L)	(° C.)	(V)	treatments	(μm)

4-1	phosphoric acid	0.3	0	195	8	0.2
4-2	phosphoric acid	0.3	7	130	8	0.2
4-3	phosphoric acid	1.0	7	80	7	0.2
4-4	phosphoric acid	1.0	20	16	1	0.2
4-5	oxalic acid	0.3	16	40	5	0.2
4-6	oxalic acid	0.3	25	40	5	0.2
4-7	sulfuric acid	0.3	0	40	5	0.2
4-8	sulfuric acid	0.3	5	25	3	0.2
4-9	sulfuric acid	0.3	16	25	3	0.4

(5) Pore Widening Treatment

Pore widening treatment was carried out by immersing the substrate for 30 minutes in an aqueous solution of phosphoric acid having a concentration of 50 g/L (solution temperature, 30° C.).
(6) Sealing Treatment
One of the following sealing treatments was carried out.
(i) Sealing Treatment 1 (Method using gold colloid particles)
The substrate was immersed for 1 minute in a gold colloid particle dispersion (average size of gold colloid particles, 120 nm) obtained by adding 1.5 mL of an aqueous solution containing 1 wt % citric acid to 1.5 mL of an aqueous solution containing 0.05 wt % HAuCl₄, using an alcohol lamp to gradually heat the mixture from room temperature, stopping heat application when the mixture turned red-violet in color, then cooling to room temperature. Following immersion, the substrate was rinsed with water and dried.
(ii) Sealing Treatment 2 (Gold Electrodeposition Method)

The substrate was immersed in a plating bath (pH 10; bath temperature, 75° C.) having the bath composition indicated in Table 5, and electrodeposition was carried out for 5 to 6 minutes using a high-purity platinum plate as the counterelectrode and at a current density of 0.1 A/dm².

TABLE 5


		Bath
Bath composition	pH	temperature

Potassium dicyanoaurate (I), 6 g/L
Potassium cyanide, 13 g/L	10	75° C.
Potassium hydroxide, 11 g/L
Potassium borohydride, 22 g/L

(iii) Sealing Treatment 3 (Copper Deposition Method)

The substrate was immersed in a dispersion (dispersion temperature, 40° C.; pH 10.2) containing 20 g/L of cuprous cyanide, 14 g/L of copper, 27 g/L of NaCN and 34 g/L of KCN, and electrodeposition was carried out for 5 to 6 minutes using a high-purity electrolytic copper plate as the counterelectrode and at a current density of 1.0 A/dm².
(iv) Sealing Treatment 4 (Nickel Deposition Method)
The substrate was immersed in a dispersion (dispersion temperature, 40° C.; pH 3) containing 220 g/L of nickel sulfate, 30 g/L of nickel chloride and 30 g/L of boric acid, and electrodeposition was carried out for 5 to 6 minutes using a high-purity nickel plate as the counterelectrode and at a current density of 2.0 A/dm².
(v) Sealing Treatment 5 (Chromium Deposition Method)
The substrate was immersed in a dispersion (dispersion temperature, 50° C.; pH 2) containing 70 g/L of anhydrous chromic acid, 0.7 g/L of sulfuric acid and 0.7 g/L of Na₂SiF₆, and electrodeposition was carried out for 5 to 6 minutes using a high-purity lead plate as the counterelectrode and at a current density of 30.0 A/dm².
(vi) Sealing Treatment 6 (Zinc Deposition Method)
The substrate was immersed in a dispersion (dispersion temperature, 20° C.; pH 5) containing 31 g/L of ZnCl₂, 15 g/L of zinc, 210 g/L of KCl and 25 g/L of boric acid, and electrodeposition was carried out for 5 to 6 minutes using a high-purity platinum plate as the counterelectrode and at a current density of 1.0 A/dm².
(vii) Sealing Treatment 7 (Silver Deposition Method)
The substrate was immersed in a dispersion (dispersion temperature, 20° C.; pH 11) containing 30 g/L of AgCN, 24 g/L of silver, 50 g/L of KCN and 15 g/L of K₂CO₃, and electrodeposition was carried out for 5 to 6 minutes using carbon as the counterelectrode and at a current density of 0.5 A/dm².
(viii) Sealing Treatment 8 (Platinum Deposition Method)
The substrate was immersed in a dispersion (dispersion temperature, 65° C.; pH 2) containing 20 g/L of platinum and 300 g/L of hydrochloric acid (using H₂PtCl₆as the reagent), and electrodeposition was carried out for 5 to 6 minutes using a high-purity platinum plate as the counterelectrode and at a current density of 15.0 A/dm².
(ix) Sealing Treatment 9 (Palladium Deposition Method)
The substrate was immersed in a dispersion (dispersion temperature, 43° C.; pH 9) containing 4 g/L of palladium, 90 g/L of ammonium nitrate and ammonia (as a pH-adjusting reagent), and electrodeposition was carried out for 5 to 6 minutes using a high-purity platinum plate as the counterelectrode and at a current density of 0.5 A/dm².
(8) Surface Treatment
Surface treatment was carried by one of the following delayering methods A to C.
(i) Delayering Method A (Mechanical Polishing Method)
Using a diamond slurry (#2,400,000; particle size, 0.1 μm) and aluminum oxide (particle size, 0.05 μm) in this order, polishing was carried out with an ultraprecision polishing system (MA-200D, manufactured by Musashino Denshi K.K.) for respectively 30 seconds and 10 seconds at a rotating speed of 50 rpm.
These polishing operations were carried out while suitably applying as a spray an aqueous solution containing 3 wt % phosphoric acid. The polishing completion time was set based on the point at which, under visual examination of the surface being-polished, the filled metal changed in color from the gloss color of the metal. For example, when the filled metal was gold, polishing was stopped when the metal changed from a gold color to a red-violet color.
(ii) Delayering Method B (Chemical Dissolution)

B-1: The surface layer was dissolved by immersing the substrate for 15 minutes (900 seconds) in an aqueous solution of 5 wt % phosphoric acid having a solution temperature of 30° C.
B-2: The surface layer was dissolved by immersing the substrate for 5 minutes (300 seconds) in an aqueous solution of chromic acid (85 wt % phosphoric acid, 118 g; anhydrous chromic acid, 30 g; pure water, 150 g) having a solution temperature of 30° C.

In both cases, the surface being subjected to dissolution was visually examined and dissolution was carried out until such time as the filled metal changed in color from the gloss color thereof.
(iii) Delayering Method C (Ion Beam Delayering in a Vacuum)
Delayering was carried out by irradiation using a flat ion milling system (E-3200, manufactured by Hitachi High-Technologies Corporation) at an irradiation surface area of 5 mm (dia.) and an ion gun voltage of 5 kV, and under a vacuum of 3×10⁻⁴Pa, while visually examining the surface using the optical microscope provided with the system.
Irradiation was carried out until the filled metal changed in color from its gloss color.
(9) Electrodeposition
The metal particles shown in Table 1 were electrodeposited by similar methods as in above-described Sealing Treatments 2 to 9. The electrodeposition time and the average size of the metal particles are shown in Table 1.
Here, the average size of the metal particles was determined by examining the surface of the structure after the electrodeposition step with a FE-SEM microscope (S-800, manufactured by Hitachi, Ltd.), measuring the diameters of 100 particles in an appropriate region, and computing the average thereof.
In the resulting structures, the average diameter of the micropores and the coefficient of variation of the pore diameter were measured by the image analysis of SEM surface micrographs obtained using a FE-SEM microscope (S-800, manufactured by Hitachi, Ltd.). Image analysis was carried out as described below.
Using image processing software (Image Factory, a product of Asahi Hi-Tech Co., Ltd.), the image was binarized (Otsu's method), following which shape analysis of the binarized image was carried out in the following order: black hole filling, black expansion, and black contraction. Next, the length shown in the micrograph was input using a measuring bar. In addition, shape characteristics were extracted and the black count and equivalent circuit diameter were output, the average pore density was computed from the black count, and the average pore diameter and standard deviation were computed from the equivalent circle diameter distribution. Moreover, the standard deviation was divided by the average pore diameter, and the coefficient of variation in pore diameter was determined.
The results are shown in Table 6.
To determine the average center-to-center spacing between neighboring micropores (pore period), each of the resulting structures was examined with a FE-SEM microscope (S-800, manufactured by Hitachi, Ltd.), the center-to-center spacing between neighboring micropores was measured for 100 pores in a suitable region, and the average of the measured values was computed.
The results are shown in Table 6.
Also, the thickness (nm) of the anodized layer following delayering by the above-described surface treatments (Delayering Methods A to C) and the removal ratio (%), which expresses what percent of the thickness prior to delayering remains, were measured as described below using a FE-SEM microscope. In addition, the average surface roughness R_a(nm) of the anodized layer after delayering was measured under the conditions indicated below using an atomic force microscope (AFM). The results are shown in Table 6.
(1) Measurement of Anodized Layer Thickness after Delayering, and Removal Ratio:
Portions of the samples before and after delayering were cut away, each portion was bent 90 degrees, and the fracture plane where the specimen cracked was examined with a FE-SEM microscope (S-800, manufactured by Hitachi, Ltd.) at a suitable magnification. The thickness of the anodized layer was measured, and the percent of the layer thickness removed was computed from the thickness of the layer before and after delayering.
(2) Measurement of Average Surface Roughness by Atomic Force Microscopy:

The average surface roughness (R_a) of the anodized layer following surface treatment was measured under the following conditions using a cantilever (SI DF40P, manufactured by Seiko Instruments, Inc.). When there were many unfilled micropores, R_awas high; when substantially all micropores were filled, R_awas low.



	Scanning area:	3,000 nm
	Scanning frequency:	0.5 Hz
	Amplitude decay rate:	−0.16
	I gain:	0.0749
	P gain:	0.0488
	Q curve gain:	2.00
	Voltage for vibration:	0.044 V
	Resonance frequency:	318.5 kHz
	Measurement frequency:	318.2 kHz
	Oscillation amplitude:	0.995 V
	Q value:	close to 460

2. Measurement of Raman Enhancement Effect

An aqueous solution containing 3×10⁻⁷mol/L of Rhodamine 6G (available as a reagent from Kanto Chemical Co., Inc.) and an aqueous solution containing 0.1 mol/L NaCl (available as a reagent from Kanto Chemical Co., Inc.) were applied to the surface of each structure obtained, following which the Raman scattering intensity at 1660 cm⁻¹was measured using a Raman spectrophotometer (T64000, manufactured by Horiba, Ltd.) at an excitation wavelength of 488 nm and a Raman shift measurement range of 1800 to 800 cm^−1.
The Raman enhancement effect was evaluated by dividing the measured Raman scattering intensity by the Raman scattering intensity at 1660 cm⁻¹measured using an ordinary slide glass with the laser output set to the maximum. When a high sensitivity was achieved, the laser output was lowered, the aqueous solution of Rhodamine 6G was diluted with water, and the enhancement factor was calculated. The results are shown in Table 6.

Indications in Table 6 of the degree of Raman enhancement are explained below.



	Excellent:	Enhancement factor of at least 10⁶
	Very good:	Enhancement factor of at least 10⁵but less than 10⁶
	Good:	Enhancement factor of at least 10⁴but less than 10⁵
	Fair:	Enhancement factor of at least 10²but less than 10³
	Poor	Enhancement factor of less than 10¹

3. Reproducibility of Signal Intensities

Ten samples for measuring the Raman enhancement effect were fabricated, the coefficient of variation in the Raman scattering intensities measured for each was computed, and the reproducibility of the Raman enhancement effect was evaluated. The coefficient of variation is defined by the following formula. The measurement results are shown in Table 6.

Coefficient of variation in Raman scattering intensity %=[(Standard deviation of Raman scattering intensity)/(average Raman scattering intensity)]×100

TABLE 6


			Thickness of
	Pore size		anodized
	coefficient		layer
Average	of	Pore	after	Percent		Raman
pore size	variation	period	delayering	removed	Ra	enhancement
(nm)	(%)	(nm)	(nm)	(%)	(nm)	effect	Reproducibility

Example 1	600	5	1200	100	50	10	Excellent	3
Example 2	350	5	700	80	80	10	Excellent	3
Example 3	350	5	700	160	60	10	Excellent	3
Example 4	350	5	700	240	40	10	Excellent	3
Example 5	350	5	700	320	20	10	Excellent	3
Example 6	120	8	240	100	50	10	Excellent	3
Example 7	40	10	120	140	30	20	Excellent	8
Example 8	30	15	100	140	30	20	Excellent	8
Example 9	30	15	100	140	30	20	Excellent	8
Example 10	15	15	60	100	50	30	Excellent	10
Example 11	10	20	50	100	50	5	Good	3
Example 12	350	5	700	240	40	10	Excellent	3
Example 13	350	5	700	320	20	10	Excellent	3
Example 14	350	5	700	240	40	10	Excellent	20
Example 15	350	5	700	320	20	10	Excellent	10
Example 16	350	5	700	240	40	10	Excellent	3
Example 17	350	5	700	320	20	10	Excellent	5
Example 18	350	5	700	240	40	10	Excellent	8
Example 19	350	5	700	320	20	10	Excellent	8
Example 20	350	5	700	240	40	10	Good	15
Example 21	350	5	700	320	20	10	Good	10
Example 22	350	5	700	240	40	10	Fair	20
Example 23	350	5	700	320	20	10	Fair	20
Example 24	350	5	700	240	40	10	Excellent	3
Example 25	350	5	700	320	20	10	Excellent	10
Example 26	350	5	700	240	40	10	Good	15
Example 27	350	5	700	320	20	10	Good	15
Comp. Ex. 1	600	5	1200	200	0	50	Very good	15
Comp. Ex. 2	350	5	700	200	0	50	Very good	15

As is apparent from Table 6, the strength and reproducibility of the Raman enhancement effect in the structures of the invention were both excellent.

Claims

1. A process for manufacturing a structure at least a portion of which is an aluminum member having on a surface thereof an anodized layer having a plurality of micropores, the process comprising the steps of, in order:

(a) anodizing a surface of an aluminum member so as to form an anodized layer having micropores present therein;

(b) sealing the micropores in the anodized layer by filling the micropores with metal;

(c) surface-treating the sealed anodized layer so as to remove at least a portion of upper layer surface thereof and set the average surface roughness (R_a) to at most 30 nm; and

(d) subjecting the surface-treated anodized layer to electrodeposition so as to form metal particles on the metal filled into the micropores during sealing.

2. The process of claim 1, wherein 20 to 80% of the anodized layer thickness is removed in the surface treatment step (c).

3. The process of claim 1, wherein the surface treatment step (c) is carried out by at least one method selected from the group consisting of mechanical polishing, chemical dissolution, and ion beam delayering in a vacuum.

4. The process of claim 3, wherein the mechanical polishing method is chemical mechanical polishing.

5. A structure obtained by the process of claim 1, wherein the micropores present in the anodized layer have an average diameter of 10 to 500 nm and a coefficient of variation in diameter of 5 to 20%, and the metal particles formed by electrodeposition have an average diameter larger than the average pore diameter yet smaller than an average center-to-center spacing between neighboring micropores.

6. A structure obtained by the process of claim 2, wherein the micropores present in the anodized layer have an average diameter of 10 to 500 nm and a coefficient of variation in diameter of 5 to 20%, and the metal particles formed by electrodeposition have an average diameter larger than the average pore diameter yet smaller than the average center-to-center spacing between neighboring micropores.

7. A structure obtained by the process of claim 3, wherein the micropores present in the anodized layer have an average diameter of 10 to 500 nm and a coefficient of variation in diameter of 5 to 20%, and the metal particles formed by electrodeposition have an average diameter larger than the average pore diameter yet smaller than the average center-to-center spacing between neighboring micropores.

8. A structure obtained by the process of claim 4, wherein the micropores present in the anodized layer have an average diameter of 10 to 500 nm and a coefficient of variation in diameter of 5 to 20%, and the metal particles formed by electrodeposition have an average diameter larger than the average pore diameter yet smaller than the average center-to-center spacing between neighboring micropores.