Porous and nanoporous semiconductor materials and their manufacture

Abstract

Methods for forming porous or nanoporous semiconductor materials are described. The methods allow for the formation of arrays of pores or nanopores in semiconductor materials with advantageous pore size, spacing, pore volume, material thickness, and other aspects. Porous and nanoporous materials are also provided.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit under 35 USC §119(e) of U.S. Provisional Application Serial No. 62 / 653,266, filed April 5, 2018, the contents of which are incorporated herein by reference in their entirety.

[0003] This application also incorporates by reference in its entirety U.S. patent application serial number 15 / 462,620, filed on March 17, 2017, and published on September 21, 2017 as U.S. Patent Publication No. US 2017 / 0271459. Technical Field

[0004] Aspects described herein generally relate to porous and / or nanoporous semiconductor materials and related methods and applications. Background Art

[0005] The production of nanoporous semiconductor materials is important for many current and potential applications, including nanofiltration, thermoelectric materials, battery electrodes, photovoltaics, and catalysis. In each of these and other applications, nanoporous semiconductors with reduced pore size, reduced interpore spacing, and increased pore aspect ratio have been found to be advantageous. However, despite current advances in nanofabrication technology, nanoporous semiconductor materials are approaching the limits of the accessible parameter space for these design variables.

[0006] Therefore, there is a need for improved methods for producing nanoporous semiconductor materials. Summary of the Invention

[0007] The present disclosure relates to the synthesis of nanoporous semiconductor materials. Certain embodiments relate to a synthesis technique utilizing metal-assisted chemical etching.

[0008] In one embodiment, the porous semiconductor material comprises a semiconductor material and a plurality of pores in the semiconductor material. The plurality of pores has an average pore size of less than 20 nm, and the plurality of pores defines a total volume porosity of at least 0.1%, as measured by dividing the total pore volume by the total pore volume plus the volume of the solid material. At least 0.05% of the pores extend through the material from one surface to an opposite or different surface, and the material has a thickness that is the smallest cross-section of the material, and the thickness is at least 0.05 microns.

[0009] In another embodiment, a method for forming a nanoporous patterned material includes covering a portion of the material with a nanoporous semiconductor membrane, and etching a portion of the material through the nanoporous semiconductor membrane. The nanoporous semiconductor membrane includes pores having an average pore size of less than 20 nm and an average aspect ratio greater than 500:1.

[0010] Other advantages and novel features of the present invention will become apparent from the following detailed description of several non-limiting embodiments of the present invention when considered in conjunction with the accompanying drawings. In the event that this specification and the documents incorporated by reference include contradictory and / or inconsistent disclosures, this specification shall prevail. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Non-limiting embodiments will be described by way of example with reference to the accompanying drawings, which are schematic and not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown is generally represented by a single numeral. For clarity, not every component is labeled in every figure, nor is every component of every embodiment of the invention shown where an illustration is not necessary to enable one of ordinary skill in the art to understand the invention. In the drawings:

[0012] Figure 1A is a reproduction of an SEM image of a crystalline Si substrate with drop-cast bare 50 nm gold nanoparticles before MACE;

[0013] Figure 1B is with Figure 1A Reproduction of an SEM image of a similar Si substrate after MACE as shown in ;

[0014] Figure 1C It is a schematic diagram of the conventional MACE process;

[0015] Figure 1D is a reproduction of an SEM image of a crystalline Si substrate with drop-cast 10 nm core SiO2-AuNPs before MACE according to one embodiment;

[0016] Figure 1E is with Figure 1D Reproduction of an SEM image of a similar crystalline Si substrate after MACE, showing a nanoporous Si material according to one embodiment;

[0017] Figure 1F is a schematic diagram of an improved MACE process according to some embodiments;

[0018] Figure 2A is a reproduction of a TEM image of 5 nm SiO2-AuNPs drop-cast on a Si substrate according to one embodiment;

[0019] Figure 2B is a reproduction of a TEM image of 10 nm SiO2-AuNPs drop-cast on a Si substrate according to one embodiment;

[0020] Figure 2Cis a reproduction of an SEM image of 5 nm SiO2-AuNPs drop-cast on a Si substrate according to one embodiment;

[0021] Figure 2D is a reproduction of an SEM image of 10 nm SiO2-AuNPs drop-cast on a Si substrate according to one embodiment;

[0022] Figure 3A is a reproduction of an SEM image of drop-cast 5 nm SiO2-AuNP catalyzed MACE for 15 minutes according to one embodiment;

[0023] Figure 3B is a reproduction of an SEM image of drop-cast 5 nm SiO2-AuNP catalyzed MACE for 30 minutes according to one embodiment;

[0024] Figure 3C is a reproduction of an SEM image of drop-cast 5 nm SiO2-AuNP catalyzed MACE for 60 minutes according to one embodiment;

[0025] Figure 3D is a reproduction of an SEM image of drop-cast 10 nm SiO2-AuNP catalyzed MACE for 15 minutes according to one embodiment;

[0026] Figure 3E is a reproduction of an SEM image of drop-cast 10 nm SiO2-AuNP catalyzed MACE for 30 minutes according to one embodiment;

[0027] Figure 3F is a reproduction of an SEM image of drop-cast 10 nm SiO2-AuNP catalyzed MACE for 60 minutes according to one embodiment;

[0028] Figure 4A is a reproduction of an SEM image of NPSi produced using drop-cast 5 nm SiO2-AuNPs via a modified mace for 60 minutes according to one embodiment;

[0029] Figure 4B yes Figure 4A Histogram of Au core size distribution and pore size distribution of NPSi material shown in;

[0030] Figure 4C is a reproduction of an SEM image of NPSi produced using drop-cast 10 nm SiO2-AuNPs via a 60-minute modified mace according to one embodiment;

[0031] Figure 4D yes Figure 4C Histogram of Au core size distribution and pore size distribution of NPSi material shown in;

[0032] Figure 5A is a reproduction of an SEM cross-sectional image of FIB-milled NPSi produced via a modified MACE process using 5 nm SiO2-AuNPs according to one embodiment;

[0033] Figure 5B is a reproduction of an SEM cross-sectional image of FIB-milled NPSi produced via a modified MACE process using 10 nm SiO2-AuNPs according to one embodiment;

[0034] Figure 5C It shows Figure 5B Figure 1 shows the EDS analysis of the SEM image.

[0035] Figure 6A is a reproduction of an SEM cross-sectional image of FIB-milled NPSi produced via a modified MACE process with 5 nm SiO2-AuNPs and an etching time of 30 minutes according to one embodiment;

[0036] Figure 6B is a reproduction of an SEM cross-sectional image of FIB-milled NPSi produced via a modified MACE process with 5 nm SiO2-AuNPs and an etching time of 60 minutes according to one embodiment;

[0037] Figure 7A is a reproduction of an SEM image of an NPSi sample produced via a modified MACE process with drop-cast 5 nm SiO2-AuNPs and an etching time of 30 minutes, immediately after etching, according to one embodiment;

[0038] Figure 7B Is making Figure 7A Reproduction of the SEM image of the NPSi sample after it was in air for two months;

[0039] Figure 8A is a reproduction of an SEM image of sputtered Ag islands on a Si substrate according to one embodiment;

[0040] Figure 8B is a reproduction of a TEM image of sputtered Ag islands on a Si substrate according to one embodiment;

[0041] Figure 8C It is shown by Figure 8B Figure 3 is a histogram of the Ag island size calculated from the TEM image;

[0042] Figure 9A is a reproduction of an SEM image of sputtered Au islands on a Si substrate according to one embodiment;

[0043] Figure 9Bis a reproduction of a TEM image of sputtered Au islands on a Si substrate according to one embodiment;

[0044] Figure 9C It is shown by Figure 9B Figure 3 is a histogram of Au island sizes calculated from TEM images;

[0045] Figure 10A It is shown by Figure 8B A histogram of the spacing between Ag islands calculated from a TEM image;

[0046] Figure 10B It is shown by Figure 9B A histogram of the spacing between Au islands calculated from a TEM image;

[0047] Figure 11A is a reproduction of an SEM image of NPSi etched with sputtered Au island catalyst according to one embodiment;

[0048] Figure 11B is a reproduction of an SEM image of NPSi etched with sputtered Ag island catalyst according to one embodiment;

[0049] Figure 11C is a reproduction of an SEM cross-sectional image of NPSi etched with sputtered Au island catalyst according to one embodiment;

[0050] Figure 11D is a reproduction of an SEM cross-sectional image of NPSi etched with sputtered Ag island catalyst according to one embodiment;

[0051] Figure 12A is a schematic diagram of a method for preparing a planar slice of an NPSi sample according to one embodiment;

[0052] Figure 12B is a schematic diagram of a TEM view of a planar slice according to one embodiment;

[0053] Figure 12C is a reproduction of a TEM image of a planar thin-sheet sample of native Si that was not exposed to a MACE process according to one embodiment;

[0054] Figure 12D is a reproduction of a TEM image of a planar sample of NPSi etched with a sputtered Au catalyst according to one embodiment;

[0055] Figure 13 is a graph showing the reflectance spectrum of NPSi etched with sputtered Au and Ag catalyst and coated with TiO2 according to one embodiment.

[0056] Figure 14Ais a reproduction of an SEM cross-sectional image of NPSi etched with sputtered Au islands and coated with TiO2 according to one embodiment;

[0057] Figure 14B is a diagram showing a method according to an embodiment Figure 14A Figure 1 shows the EDS spectra of the SEM images at two different points.

[0058] Figure 15 Graph showing XPS analysis of ALD-coated NPSi according to one embodiment;

[0059] Figure 16A An optical microscope image showing five crystalline Si windows on a single silicon mask chip according to one embodiment;

[0060] Figure 16B TEM image showing a Si window coated with Ag nano-island catalyst according to one embodiment;

[0061] Figure 16C shows a TEM image of a porous Si window after 1 minute of MACE according to one embodiment;

[0062] Figure 16D Showing a 15 μm thickness of about 1 cm according to one embodiment 2 Photo of the NPSi mask;

[0063] Figure 16E Show Figure 16D SEM cross-sectional image of a 15 μm thick NPSi mask;

[0064] Figure 16F Show Figure 16E SEM image of the bottom surface of the NPSi mask;

[0065] Figure 17A is a schematic diagram of an NPSi-assisted patterning process for a nanoporous monolayer material and a subsequent thermal annealing process for controlling porosity according to one embodiment;

[0066] Figure 17B shows an SEM image of a pristine MoS2 domain on a Si / SiO2 substrate according to one embodiment;

[0067] Figure 17C shows an SEM image of a single triangular MoS2 domain after undergoing a patterning process according to one embodiment;

[0068] Figure 17D shows an SEM image of nanoporous MoS2 domains produced by selective etching through a thin NPSi mask according to one embodiment;

[0069] Figure 17E Shown from Figure 17D SEM images of star-shaped nanoporous MoS2 domains on the same substrate as shown in;

[0070] Figure 17F Depicts Figure 17E Magnified view of the area indicated by the box in ;

[0071] Figure 17G shows a high magnification SEM image of a nanoporous MoS2 domain according to one embodiment;

[0072] Figure 17H is an SEM image of a nanoporous WS2 domain after O2 plasma etching using an NPSi mask according to one embodiment;

[0073] Figure 17I yes Figure 17H Magnified view of the sample shown in;

[0074] Figure 17J yes Figure 17H Another enlarged view of the sample shown in;

[0075] Figure 18A shows an SEM image of MoS2 domains after O2 plasma treatment through a NPSi mask according to one embodiment;

[0076] Figure 18B Show Figure 18A SEM image of MoS2 domains after thermal annealing in air for 30 minutes;

[0077] Figure 18C Shown from Figure 18B SEM image of the MoS2 domain of the same sample after 60 minutes of annealing;

[0078] Figure 18D Shown from Figure 18B SEM image of the MoS2 domain of the same sample after 100 minutes of annealing;

[0079] Figure 18E Show Figure 18A A magnified view of a sample of FIG, and showing independent pores with a diameter of about 70 nm;

[0080] Figure 18F Show Figure 18B A magnified view of a sample of FIG201 and showing the nanopattern of enlarged pores produced in a single layer of MoS2;

[0081] Figure 18G Show Figure 18C A magnified view of the sample, and showing further expansion and convergence of the pores;

[0082] Figure 18H Show Figure 18D A magnified view of a sample of FIG. 1 and showing MoS2 flakes and oxidized MoS2 particles;

[0083] Figure 18I is a graph of MoS 2 edge density and porosity versus thermal annealing time for an O 2 plasma etched sample according to one embodiment.

[0084] Figure 18J shows an SEM image of a MoS domain half on a NPSi etch mask and half on non-porous silicon after O plasma treatment and 60 minutes of thermal annealing according to one embodiment;

[0085] Figure 18K shows an SEM image of a pristine MoS2 domain after thermal annealing in air for 100 minutes according to one embodiment;

[0086] Figure 19A is a graph of the normalized photoluminescence spectrum of a pristine MoS2 domain after thermal annealing according to one embodiment;

[0087] Figure 19B is a graph of normalized PL spectra of nanoporous MoS2 domains after thermal annealing for up to 70 minutes according to one embodiment;

[0088] Figure 19C yes Figure 19B Figure 2 shows the normalized PL spectrum of an example nanoporous MoS2 domain after continued thermal annealing.

[0089] Figure 19D is a graph of the normalized Raman spectrum of a pristine MoS2 domain after thermal annealing according to one embodiment;

[0090] Figure 19E is a graph of a normalized Raman spectrum of a nanoporous MoS2 domain after thermal annealing according to one embodiment;

[0091] Figure 19F is a graph showing the relative maximum PL intensity of nanopatterned MoS2 domains and pristine MoS2 domains as a function of annealing time according to one embodiment;

[0092] Figure 20A is a graph of normalized PL spectra of pristine MoS2 domains before and after direct O2 plasma treatment according to one embodiment;

[0093] Figure 20B is a graph of normalized Raman spectra of pristine MoS2 domains before and after direct O2 plasma treatment according to one embodiment;

[0094] Figure 20C shows an SEM image of a pristine MoS2 domain after 1 second of direct O2 plasma treatment according to one embodiment;

[0095] Figure 20D Show Figure 20C SEM images of examples at higher magnification;

[0096] Figure 20E Show Figure 20D SEM images of examples at higher magnification;

[0097] Figure 20F shows an SEM image of a pristine MoS2 domain after 10 seconds of direct O2 plasma treatment according to one embodiment;

[0098] Figure 20G Show Figure 20F SEM images of examples at higher magnification;

[0099] Figure 20H Show Figure 20G SEM images of examples at higher magnification;

[0100] Figure 21A is a schematic diagram of a method for producing an NPSi membrane according to one embodiment;

[0101] Figure 21B Photographs showing the front and back surfaces of an original film according to one embodiment;

[0102] Figure 21C shows a SEM of an initially etched surface according to one embodiment;

[0103] Figure 21D Show Figure 21C SEM image of the surface opposite to the initial etched surface shown in;

[0104] Figure 21E is a cross-sectional SEM image of a film according to one embodiment;

[0105] Figure 21F is a graph of BJH pore size distribution calculated from BET nitrogen adsorption and desorption experiments according to one embodiment;

[0106] Figure 22A is an image of static water contact angle on a pristine film according to one embodiment;

[0107] Figure 22B is an image of static water contact angle on a piranha treated film according to one embodiment;

[0108] Figure 23 is a graph of permeate mass versus time during pure water and 5 nm AuNP rejection experiments according to one embodiment;

[0109] Figure 24A is a graph of permeability and retention during a 5 nm AuNP filtration experiment according to one embodiment;

[0110] Figure 24B Shown in Figure 24A Photos of the samples collected during the experiment;

[0111] Figure 24C Shown from Figure 24A SEM image of the surface of the membrane of the example after the test;

[0112] Figure 24D Shown from Figure 24A Cross-sectional SEM images of the membrane of the example;

[0113] Figure 25A A graph showing pure water permeability and rejection of two molecular dyes for two filtration membranes according to one embodiment;

[0114] Figure 25B Shown for use from Figure 25A Photographs of feed (left) and permeate (right) samples of a molecular dye filtered through a membrane.

[0115] Figure 25C Shows the use of Figure 25B Photographs of feed (left) and permeate (right) samples of a second molecular dye filtered through the same filter membrane.

[0116] Figure 25D Shown for use from Figure 25A The second filter membrane of the example is filtered with Figure 25B Photographs of feed (left) and permeate (right) samples of the same molecular dye.

[0117] Figure 25E Shows the use of Figure 25D The same membrane filtration Figure 25C Photographs of feed (left) and permeate (right) samples of the same molecular dye.

[0118] Figure 26A A photograph showing feed permeate for an oil-water separation test according to one embodiment;

[0119] Figure 26B Shown from Figure 26A Concentrate of oil-water separation test examples;

[0120] Figure 26C It is shown in Figure 26A Photograph of hexadecane being absorbed from the membrane surface after the experiment;

[0121] Figure 26D It shows Figure 26A Graphs of TGA measurements of examples;

[0122] Figure 26E Shown from Figure 26A Optical microscope image of a 50-fold diluted feed of the example;

[0123] Figure 26F Shown from Figure 26A Optical microscope images of undiluted permeates of examples;

[0124] Figure 26G A photograph showing solid hexadecane recovered during a ZLD test according to one embodiment;

[0125] Figure 26H Shown in Figure 26G Another photograph of solid hexadecane recovered during a ZLD test of Example;

[0126] Figure 26I Shown in Figure 26G Another photograph of solid hexadecane recovered during a ZLD test of Example;

[0127] Figure 26J yes Figure 26G Graph of TGA measurements of ZLD tests of examples;

[0128] Figure 26K is shown for Figure 26G A graph of the droplet size distribution of an emulsified hexadecane emulsion feed solution measured using DLS for an example ZLD test;

[0129] Figure 27 is a cross-sectional SEM image of catalyst deposited directly on opposite sides of a thick Si wafer according to one embodiment; and

[0130] Figure 28 is a graph of pore size distribution for a 400 μm thick membrane according to one embodiment. DETAILED DESCRIPTION

[0131] The present inventors have recognized and understood that current nanofabrication techniques are not always well suited for producing nanoporous semiconductor materials in a scalable manner while maintaining the ability to finely control the morphology of the resulting nanoporous semiconductor materials. For example, it has recently been demonstrated that electron beam lithography combined with deep reactive ion etching of the current prior art can produce nano barrel structures with a wall thickness of 6.7 nm and an aspect ratio of 50:1. Although potentially useful for manufacturing specific components in nanoelectronics, the extremely high cost and long processing time of electron beam lithography limit this technology to device sizes of square microns, making it an unsuitable technology for any of the above-mentioned applications requiring larger device sizes. Block copolymer lithography has attracted considerable attention as a more scalable alternative to electron beam lithography, and has recently been combined with plasma etching to produce sub-10 nm features with an aspect ratio of 17:1. However, both techniques still require ion bombardment of the substrate in a vacuum, and therefore are inherently limited in their ability to be integrated into a high-throughput manufacturing process (e.g., the process required for the previously mentioned applications).

[0132] Solution-based technology has also been explored as an alternative to photolithography-based technology. For example, metal-assisted chemical etching (MACE) is an electrochemical technique that relies on the anisotropic etching of precious metal catalysis of nanopores in semiconductor materials via a simple, scalable and low-cost solution-based process. For example, for a gold catalyst deposited on the surface of a silicon (Si) substrate and placed in an aqueous solution of hydrofluoric acid (HF) and hydrogen peroxide (H2O2), the general reaction mechanism can be explained as follows. First, H2O2 is reduced at the surface of the nanoparticles that constitute the cathode reaction. Holes (h+) are generated in this reduction and diffuse from the particles to the Si substrate, which are subsequently oxidized and dissolved by HF at the anode. The entire reaction also includes protons (H+) being reduced to hydrogen, which is released as a gas (H2). As etching proceeds, the gold nanoparticles remain close to Si via van der Waals interactions, thereby continuing to catalyze the reaction.

[0133] Recently, MACE has become the focus of a large amount of work that realizes precious metal pattern in the etching of positive feature (such as nanowire) and negative feature (comprising nanopore). The common method of forming negative nanopore via MACE is included in depositing colloidal nanoparticle catalyst on semiconductor surface, depositing film and dewetting, or growing from solution. In these technologies, the maximum degree control to catalyst size, monodispersity and position is provided by drop casting or similar method deposition of pre-synthesized precious metal nanoparticles. Although this method is inherently low-cost and very scalable, compared with the pattern limited, sputtered or grown by the interconnection lithography with the interface flat with substrate, the etching mobility of nanoparticles along crystal orientation is much more difficult to control. It has been found that this causes the offset of particle in the lateral direction and the partial loss of anisotropy, causes the significant change of hole depth and direction. The reason of this phenomenon may include the non-spherical property of particle, the particle displacement caused by the hydrogen produced and the inhomogeneous injection of hole from particle to surrounding semiconductor material. This challenge is complicated for very small nanoparticles, because its shape becomes dominated by facet and no longer resembles sphere.

[0134] In view of the above, the present inventors have recognized and appreciated the numerous benefits associated with methods for producing nanoporous semiconductor materials that overcome the aforementioned disadvantages associated with conventional photolithography and MACE techniques. For example, the methods described herein can allow for the production of nanoporous semiconductor materials having smaller pore sizes, smaller inter-pore spacings, and larger pore aspect ratios compared to existing methods, while also being scalable to larger areas and device sizes.

[0135] In addition to the above, the inventors have understood that the nanoporous semiconductor materials described herein can be used in conjunction with many applications including, but not limited to, masking and patterning, membrane filtration applications, and sensing, catalysis, and in electronic devices. In particular, the inventors have understood that in many of these applications, precise control of porosity, pore size, and / or pore position may be important for obtaining desired mechanical, thermal, electrical, and / or transport properties. As discussed further below, the methods and materials described herein can allow such control of the pore characteristics of nanoporous semiconductor materials so that they can be used in the above-mentioned applications.

[0136] According to some embodiments, the method for producing nanoporous semiconductor material comprises that a plurality of nanoparticles are arranged on the surface of semiconductor substrate (for example, by drop casting of the solution of nanoparticles), and makes nanoparticles self-assemble into close-packed monolayer array via solvent evaporation.Each nanoparticle comprises the sacrificial spacer layer that surrounds smaller noble metal nanoparticle core.For example, in certain embodiments, noble metal nanoparticle core can comprise gold, silver, platinum and / or palladium, and sacrificial spacer layer can be oxide, for example silicon dioxide (SiO 2 ).After the semiconductor material is coated with the layer of nanoparticles, it is immersed in etching solution (for example, comprising acid such as hydrofluoric acid and MACE solution of oxidant such as hydrogen peroxide), sacrificial spacer layer is partially or completely removed, leaving the array of noble metal nanoparticles (comprising some residual sacrificial material or not comprising sacrificial material) of good spacing on the surface of semiconductor.Therefore, during deposition and self-assembly, sacrificial spacer layer keeps minimum separation between noble metal nanoparticles. The inventors have discovered that these spaced-apart noble metal nanoparticles can then be catalytically etched into the semiconductor surface to form nanopores having smaller diameters, smaller interpore spacings, and large pore aspect ratios than those achievable with conventional solution-based etching techniques. Without wishing to be bound by any particular theory, the size and interpore spacing of the etched pores can be controlled by controlling the size of the catalytic noble metal nanoparticles and the thickness of the sacrificial spacer layer, respectively.

[0137] In certain embodiments, the method for forming nanoporous semiconductor material comprises that a plurality of nanoparticles containing precious metals are arranged near semiconductor substrate.As used herein, the nanoparticles arranged near semiconductor substrate generally refer to the nanoparticles arranged on the surface of adjacent semiconductor substrate, which may include at least a portion of nanoparticles in direct contact with substrate.Should be understood that any suitable method (such as drop casting, spin coating, self-assembled monolayer formation technology such as Langmuir-Blodgett groove) may be used to arrange a plurality of nanoparticles containing precious metals on the surface.

[0138] As mentioned above, in some cases, the nanoparticle containing precious metal comprises the precious metal core that is at least partially surrounded by sacrificial material (for example, sacrificial spacer).As used herein, sacrificial material generally refers to be intended to semiconductor substrate is processed with the material that is at least partially removed before forming hole therein, and sacrificial material can be removed by exposing sacrificial material to the environment (for example, solvent) that dissolves sacrificial material.For example, as mentioned above, in one embodiment, sacrificial material can comprise SiO , it can be quickly dissolved via being exposed to the HF in the MACE solution.But, should be understood that other sacrificial materials and / or solvent also may be suitable, because present disclosure has no restriction in this respect.

[0139] According to specific embodiments, the nanoparticles containing the noble metal can be nanostructures with any suitable shape, including but not limited to spherical, rod-shaped, linear, cubic, pyramidal, prism-shaped and irregular shapes. In addition, the noble metal core of the nanoparticles containing the noble metal can have a shape roughly the same as the overall shape of the nanoparticle, or the core can have a shape different from the nanoparticle. Therefore, it should be understood that for the noble metal core and / or the nanoparticles containing the noble metal, the present disclosure is not limited to any specific shape and / or configuration.

[0140] In some embodiments, a plurality of nanoparticles containing noble metals can be assembled into arrays, such as close-packed arrays of nanoparticles. As used herein, arrays generally refer to at least partially ordered patterns, such as two-dimensional patterns, in which at least a portion of a plurality of nanoparticles has similar spacing relative to each other. Closely packed arrays refer to arrays in which at least a portion of a nanoparticle is in direct contact with two or more of its nearest neighbor's nanoparticles. For example, in hexagonal close-packed arrays, each nanoparticle can be in direct contact with six nearest neighbor's nanoparticles. In some embodiments, a plurality of nanoparticles containing noble metals can be arranged with sacrificial materials of close-packed arrays and directly contacted adjacent nanoparticles. After removing the sacrificial material, the noble metal core can remain in the array of non-close packing (i.e., arrays at intervals). As described above, a plurality of nanoparticles can self-assemble to form arrays, i.e., after nanoparticles are arranged near a semiconductor substrate, nanoparticles can naturally arrange themselves into ordered patterns. In some cases, self-assembly can be driven by the evaporation of a solvent, such as wherein dispersed with an aqueous solution of nanoparticles.

[0141] According to another embodiment, the method for producing nanoporous semiconductor material is included in and forms a plurality of noble metal islands on the surface of semiconductor substrate.In some cases, forming noble metal island can comprise via suitable deposition process noble metal is deposited on semiconductor substrate, and makes island self-assembly by the interface energy of noble metal and semiconductor substrate.In some cases, the self-assembly of noble metal island can produce island of size and uniform interval.After forming noble metal island, by etching (for example, by semiconductor is immersed in MACE solution), form a plurality of holes in semiconductor substrate.Similar to the embodiment discussed above, the inventor has recognized that noble metal island can catalyze the etching of semiconductor surface to form the nanopore with less diameter, less inter-hole spacing and large hole aspect ratio compared with the nanopore that can be realized by the etching technique based on solution with conventional.

[0142] It will be understood that the noble metal islands can be formed by any suitable deposition process. For example, in some embodiments, a thin layer of noble metal can be deposited by a physical vapor deposition process (e.g., a sputtering process (e.g., magnetron sputtering), electron beam assisted evaporation, or thermal evaporation). As described above, the noble metal islands can be formed naturally by the interface energy of the noble metal-semiconductor interface. Without wishing to be bound by any particular theory, in some embodiments, the size and spacing of the noble metal islands can be controlled by appropriately controlling the surface energy of the semiconductor surface and / or one or more aspects of the deposition process (e.g., the amount of material deposited). In addition, in some cases, a specific deposition process can produce a planar interface between the noble metal island catalyst and the semiconductor substrate surface, which can allow for highly anisotropic etching behavior, which can result in holes with higher aspect ratios.

[0143] As used herein, etching generally refers to a part for chemically removing (for example, semiconductor substrate) via being exposed to etching solution.In some embodiments (for example, the embodiment utilizing MACE process), etching process may be affected by the presence of the catalyst on substrate (for example, semiconductor substrate) surface.Therefore, the method according to some embodiments can include etching the surface of substrate by the pattern affected by the array of catalyst particles (for example, the array of noble metal nanoparticles or noble metal islands formed after deposition process (for example, sputtering process)).Especially, the pattern of the features etched into substrate can be determined at least in part by the position of catalyst particles in array.In addition, it should be understood that the present disclosure is not limited to any specific catalyst particles.For example, noble metal catalyst particles (for example, nanoparticles and / or islands) are as described above, and can include noble metals, for example gold, silver, platinum, palladium etc.In addition, in some embodiments, the catalyst particles made of other metals (for example, copper) also may be suitable.

[0144] In some embodiments, the methods described herein can be used to form nanoporous semiconductor materials having pore sizes and / or interpore spacings less than 15 nm, less than 10 nm, or less than 6 nm, and the aspect ratio of the pores can be greater than 50:1, greater than 75:1, greater than 100:1, greater than 200:1, greater than 300:1, or greater than 375:1, greater than 400:1, greater than 500:1, greater than 750:1, greater than 1000:1, greater than 2500:1, greater than 5000:1, greater than 10000:1, or more. In certain embodiments, the porosity of the nanoporous semiconductor material can be greater than 0.5%, greater than 1%, greater than 5%, greater than 10%, or greater than 15%. For example, in one embodiment, the porosity can be about 18%. In some embodiments, porosity can refer to the fraction of the total surface area of ​​the semiconductor material covered by pores (i.e., the total area of ​​the pores defined by the pores divided by the total surface area of ​​the semiconductor substrate). Suitable semiconductor materials include, but are not limited to, silicon, gallium arsenide, indium phosphide, germanium, and silicon-germanium alloys; depending on the particular embodiment, the semiconductor material may be crystalline (ie, single crystal or polycrystalline).

[0145] Furthermore, it should be understood that the present disclosure is not limited to any particular etching solution. In some embodiments, a MACE solution comprising a mixture of an acid (e.g., hydrofluoric acid) and an oxidant may be suitable, and the specific acid and oxidant may be selected based on the specific semiconductor material being etched. For example, a mixture of hydrofluoric acid and hydrogen peroxide may be suitable for etching silicon, and a mixture of sulfuric acid and potassium permanganate may be suitable for etching gallium arsenide and indium phosphide.

[0146] In certain embodiments, the methods described herein may further comprise depositing a functional layer on the surface of the pores of the nanoporous semiconductor material. For example, the pore surface may be functionalized with an oxide material, such as aluminum oxide (e.g., Al2O3) or titanium oxide (e.g., TiO2) or other materials, such as nitrides. It will be appreciated that the functional layer may be deposited using any suitable deposition technique, including but not limited to atomic layer deposition (ALD) and chemical vapor deposition (CVD).

[0147] In some embodiments, the average pore size of the plurality of pores in the nanoporous semiconductor material can be less than about 10 nm, and the plurality of pores can define a total volume porosity of at least 0.5%. For example, the total volume porosity can be measured as the total pore volume in the nanoporous semiconductor material divided by the total pore volume of the nanoporous semiconductor material plus the volume of the solid material. In some embodiments, the total volume porosity can be greater than about 0.5%, greater than about 0.75%, greater than about 1%, greater than about 2%, greater than about 5%, greater than about 7%, greater than about 10%, greater than about 20%, greater than about 30%, or greater. In some cases, the total volume porosity can be less than about 50%, less than about 40%, less than about 30%, less than about 20%, less than about 10%, and / or less than about 5%.

[0148] In some embodiments, a portion of the pores formed in the nanoporous semiconductor material can extend from a first surface of the material through the material to an opposite or different surface of the material. For example, in some cases, the percentage of pores extending through the material can be at least about 0.05%, at least about 0.075%, at least about 0.1%, at least about 0.25%, at least about 0.5%, at least about 0.75%, at least about 1%, at least about 2.5%, at least about 5%, at least about 7.5%, at least about 10%, and / or at least about 20%. In other cases, the percentage of pores extending through the material can be less than about 30%, less than about 20%, less than about 10%, less than about 5%, less than about 1%, and / or less than about 0.5%.

[0149] In some embodiments, the thickness of the nanoporous material can be measured as the minimum cross-sectional thickness of the material for about 0.05 micron to about 400 microns. For example, thickness can be greater than about 0.05 micron, greater than about 0.075 micron, greater than about 0.1 micron, greater than about 0.2 micron, greater than about 0.5 micron, greater than about 0.5 micron, greater than about 1 micron, greater than about 2.5 microns, greater than about 5 microns, greater than about 10 microns, greater than about 25 microns, greater than about 40 microns, greater than about 60 microns, greater than about 75 microns, greater than about 85 microns, greater than about 100 microns, greater than about 150 microns, greater than about 200 microns, greater than about 250 microns or larger. In some cases, thickness can be less than about 400 microns, less than about 300 microns, less than about 200 microns, less than about 100 microns, less than about 50 microns, less than about 10 microns, less than about 1 micron, less than about 0.5 micron and / or less than about 0.1 micron.

[0150] As described above, in some applications, the nanoporous semiconductor materials described herein can be used in combination with masks and / or patterning applications. For example, nanoporous semiconductor materials can be used as masks to form a desired pattern (e.g., a hole pattern) on another material. In some embodiments, such a mask can be used as an etching mask to form a nanoporous structure in a two-dimensional material (e.g., molybdenum disulfide (MoS2) and / or tungsten disulfide (WS2)). For example, in an exemplary embodiment, a nanoporous silicon film material with an average pore aspect ratio greater than 1000:1 and a pore size less than 20 nm can be used as an etching mask for nanopatterning two-dimensional MoS2 and / or WS2 materials (or other suitable two-dimensional materials). The mask can have a lateral dimension in the range of 100 μm by 100 μm to 1 cm by 1 cm and a thickness in the range of 50 nm to 15 microns. Applying these masks to a two-dimensional material and then performing an etching process can produce nanopores with a diameter of about 70 nm in the two-dimensional material, and if necessary, the pores can be enlarged by thermal annealing in air. The present inventors have appreciated that this nanopatterning process can allow for control of the edge-to-area ratio of a two-dimensional material, which can allow for tuning the properties of the two-dimensional material for various applications, such as filtering, sensing, and / or electrocatalysis. For example, in catalytic applications, greatly increasing the edge density achieved by the methods and materials described herein can provide improved catalytic performance, such as in the hydrogen evolution reaction.

[0151] While specific patterning applications have been described above in conjunction with two-dimensional materials (e.g., MoS2 and WS2), it will be understood that the present disclosure is not limited to any particular type of patterning application, and that the methods and materials described herein can be applied to a wide range of patterning applications. Furthermore, while specific size and pore characteristics of nanoporous semiconductor materials have been described in conjunction with the above embodiments, it will be understood that various patterning applications can utilize nanoporous semiconductor materials having any suitable combination of size and pore characteristics, as the present disclosure is not limited in this respect.

[0152] In addition, as described above, the nanoporous semiconductor materials described herein can be used in combination with various filter membranes. The inventors have understood that the materials disclosed herein can provide many benefits relative to existing filter membranes (such as polymer membranes and / or ceramic membranes). For example, most polymer membrane technologies are only stable at temperatures below 50°C in aqueous environments and in a pH range of 4 to 10, and ceramic materials face many challenges in terms of cost and scalability required for commercialization, especially at very small filtration scales, such as less than 1 nm. In contrast, the nanoporous materials described herein may be able to separate at the sub-1 nm molecular scale and be suitable for a variety of chemical and thermally harsh environments. In addition, the methods described herein can be used to manufacture nanoporous materials, which can facilitate large-scale economically viable manufacturing.

[0153] For example, in some embodiments, nanoporous semiconductor films can be produced by first thinning a portion of the semiconductor material to achieve a desired thickness, depositing a metal catalyst on the thinned portion, and then etching nanopores through the thinned portion, for example, using the etching method described herein. In other embodiments, thicker film materials can be formed by directly etching the semiconductor material without thinning. For example, nanopores can be etched through a semiconductor material having a thickness of about 200 microns to about 400 microns. In some cases, etching can be performed from each of the two sides of the material (for example, opposite sides) by depositing an etching catalyst on each of the two sides. In view of the larger thickness of the semiconductor material, the nanopores formed in such a material may have a very large aspect ratio, for example, greater than 20,000:1, greater than 40,000:1, greater than 60,000:1, greater than 70,000:1, or higher.

[0154] Example

[0155] The following examples are intended to illustrate certain embodiments of the invention, but are not intended to illustrate the full scope of the invention.

[0156] Example 1 - Coated Nanoparticle Catalyst

[0157] In one group of illustrative embodiments, the improved MACE process based on solution is used to synthesize nanoporous silicon (NPSi) with sub-10nm pore size, sub-10nm inter-hole spacing and the hole aspect ratio more than 100:1. The method allows the ordered nanopore array of new size range to be manufactured simultaneously, increases the etching uniformity and the anisotropy of the MACE catalyzed by nanoparticles, and significantly improves the scalability and high throughput properties of the process relative to conventional photolithography MACE method. As discussed in more detail below, by first drop-casting silica shell gold nanoparticles (SiO2-AuNP) on crystalline Si substrates, simple two-step method is carried out. Then solvent evaporation promotes SiO2-AuNP self-assembly into close-packed monolayer array. Secondly, by Si immersed in MACE solution through SiO2-AuNP monolayer coating, silica shell is consumed quickly by HF, leaves the array of exposed AuNP well spaced from the surface. Then, these AuNPs seamlessly catalyze nanopore formation with etching accuracy and consistency not previously observed in sub-10nm range. The deposited SiO2-AuNP monolayer before MACE and the array of nanopores after MACE were characterized by high-resolution transmission electron microscopy (TEM) and scanning electron microscopy (SEM), respectively, allowing detailed monitoring of the pore size, interpore spacing, and pore aspect ratio of the resulting NPSi.

[0158] SiO2-AuNP (5nm and 10nm gold core) is purchased from Sigma-Aldrich USA. The 5nm gold core particles are diluted twice with H2O and then diluted five times with acetone. The 10nm gold core particles are diluted twice with acetone. Silicon wafer (VirginiaSemiconductor Inc.) is boron-doped, wherein the resistivity is 0.001Ω·cm to 0.01Ω·cm, and the thickness is 275μm+ / -25μm. The prepared solution is drop-cast on the silicon wafer and allowed to dry in air. The coated substrate is then added to MACE solution (5.65M HF, 0.12M H2O2) for different times, as described in more detail below. The sample is removed and rinsed with deionized water to stop the reaction. In order to image the pore cross section, atomic layer deposition (ALD, CambridgeNanoTech Savannah) is used to fill the pores with Al2O3 for the preservation and image contrast of the porous structure. The particle size and pore size from SEM and TEM images were analyzed using ImageJ and MATLAB. The samples were imaged using a Zeiss Ultra Plus Field Emission SEM, a FEI Helios 660 focused ion beam (for cross-section milling) with a SEM (attached with an EDAX energy dispersive X-ray spectrometer [EDS] detector), and a JEOL 2100 transmission electron microscope.

[0159] As described above, the use of noble metal nanoparticles deposited from solution to catalyze the etching of nanopores in Si via a conventional MACE process offers the benefit of being a scalable route to fabricate NPSi. Figure 1C A conventional MACE process is schematically shown, wherein a solution 110 containing colloidal nanoparticles dispersed in a solvent is first deposited on a silicon substrate 120. The solvent is allowed to evaporate, leaving AuNPs 130 on the silicon substrate 120, and then AuNP catalytic etching is performed. Figures 1A to 1C As shown, this method is accompanied by a series of challenges of its own, and first is to control nanoparticles to be arranged on the surface, and second is to guarantee that particle etches in the direction perpendicular to substrate surface during this process.The former becomes difficult owing to many complicated forces of the behavior of controlling nanoparticles in solution and in the self-assembly process during solvent evaporation.These comprise van der Waals force, electrostatic force, steric drag, solvation force, consumption power, capillary force, convection force and frictional force.The result of these interactions is that nanoparticles tend not to form the individual layer of periodic spacing usually, but show clustering behavior. Figures 1A to 1B (Scale bar: 100 nm) shows this phenomenon. Figure 1A Shown is the aggregation of bare 50 nm AuNPs 130 deposited via drop casting from solution onto a crystalline Si wafer 120. This behavior is particularly detrimental to etching nanopores via a MACE process. Figure 1B An attempt to etch a nanopore using a similar aggregated group of bare 50 nm AuNPs is shown; the aggregated group of nanoparticles resulted in the etching of a relatively large, non-uniform depression 140. The result of this attempt was the etching of a relatively large, non-uniform depression 130 in the Si (a topography of little use for most relevant applications requiring size-controlled individual pores). Figure 1C -2). Other potential undesirable consequences of conventional MACE processes include lateral etching of the nanocatalyst ( Figure 1C -3) or no etching at all ( Figure 1C -4). AuNPs are understood to preferentially <100> The etch is catalytically directed in this direction, but deviations from this path may also occur due to uneven injection of holes from other surrounding catalysts and their own facets, or displacement caused by hydrogen generated during etching.

[0160] In contrast, Figures 1D to 1F The improved MACE process according to the present disclosure is shown in FIG, which includes the self-assembly of SiO2-AuNP monolayer arrays from solution on a crystalline Si substrate. Figure 1F A solution 150 containing colloidal SiO2-AuNPs 160 is drop-cast onto a silicon substrate 120. SiO2 serves as a sacrificial spacer layer that maintains separation between the AuNP cores during deposition and self-assembly, and as shown in FIG. Figure 1D and 1F As shown, SiO2-AuNPs 160 assemble into well-spaced arrays after solvent evaporation. Upon immersion in MACE etchant, the SiO2 shell is rapidly consumed, leaving behind periodically spaced AuNPs 130, which initiate the etching process to form NPSi material 170. Figure 1D A monolayer of SiO2-AuNPs with a 10 nm gold core and a SiO2 shell of 15 nm to 20 nm thickness is depicted, while Figure 1E An etched Si substrate, NPSi material 170, is shown, which was initially coated in a similar monolayer and then exposed to a MACE solution for one hour. Figures 1D to 1E The scale bar in the figure is 100 nm.

[0161] SiO2-AuNPs can be synthesized using tetraethyl orthosilicate (TEOS) as a precursor with a simple solution chemistry process that allows for precise control of shell thicknesses ranging from 2 nm to 90 nm. In this example, gold-core SiO2-AuNPs with diameters of 5 nm and 10 nm (SiO2 shell thicknesses of 15 nm to 20 nm) were used to explore a new range of pore sizes in NPSi and achieve a fine degree of control over pore size and spacing. Figure 2A and 2B TEM images of 5 nm and 10 nm SiO2-AuNPs drop-cast onto (100) Si wafers from aqueous solutions of acetone and ultrapure water are shown, respectively; scale bar is 100 nm, and for Figure 2A and 2B The scale bar of each inset is 40 nm. Figure 2A 、 2B ) shows both an outer SiO2 shell and a well-spaced inner gold core. Some regions of the monolayer exhibit a well-ordered hexagonal close-packed pattern, while others are slightly more widely spaced. Regions of sparse bilayer formation on top of the more densely packed monolayer can also be observed ( Figure 2A The behavior of these partial bilayers during MACE is discussed in more detail below in the context of porosity, interpore spacing, and process control. Figure 2C and 2D Lower magnification SEM images of similar 5 nm and 10 nm SiO2-AuNP arrays over multiple micrometers are shown, illustrating the easy scalability of catalyst deposition. Scale bars are in Figure 2C 1 μm in the Figure 2D The scale bar in the figure is 500 nm, and the scale bar in the inset is 100 nm. Although further exploration and optimization of different deposition techniques (such as tilted assembly) are needed to achieve centimeter-scale arrays of SiO2-AuNP catalysts, there are no fundamental obstacles to their realization.

[0162] To show the time-based progression of the improved MACE process, drop-cast samples were exposed to the etchant solution for 15 minutes, 30 minutes, and 60 minutes. Figures 3A to 3C SEMs of 5 nm gold core particles are shown after 15 minutes, 30 minutes and 60 minutes, respectively. Figures 3D to 3F SEM images of 10 nm gold core particles are shown after 15 minutes, 30 minutes, and 60 minutes, respectively. Figures 3A to 3F The scale bar in each of the is 100 nm. The images in FIG3 show the progression of the MACE process after the SiO2 shell dissolves. For both particle sizes (5 nm gold core and 10 nm gold core), the SiO2 shell should dissolve almost immediately after introduction of the MACE solution, leaving an array of evenly spaced bare AuNPs. It was observed that many of the 5 nm core particles began to etch into the Si within 15 minutes ( Figure 3A ), while few 10 nm core particles showed etching in the same time interval ( Figure 3D ). After 60 minutes, for both the 5 nm and 10 nm samples, most of the particles had etched into the Si and were no longer visible on the surface. Even if ideally, all particles should be etched into the Si once their SiO2 shells dissolve, there are two possible reasons why this may not happen: (1) Once a particle's shell is etched, it may come into contact with the Si surface at the vertices of its facets or on the edge of one of its grains. Since the hole injection rate from a single particle is proportional to its contact surface area with the Si, its etching rate is significantly lower than that of particles with a planar interface between its facets and the Si surface. (2) Depending on the distribution of the SiO2 shell thickness of the particles, some particles may start etching before their closely adjacent neighbors, which may cause a small number of particles to remain on the surface due to the lack of hole injection near the surrounding particles that have already penetrated the substrate.

[0163] Since directional etching is facilitated by uniform hole injection into Si, particles lacking adjacent neighbors are less likely or not to etch downward. A promising observation is that little lateral or clustered etching occurs after 60 minutes, which can be explained by the high uniformity of hole injection from closely spaced AuNPs, which appears to facilitate the directional etching in the Si. <100> Therefore, the inventors infer that the quality of NPSi depends mainly on the quality and monodispersity of SiO2-AuNPs treated with MACE, as well as the quality of the resulting self-assembled monolayer (SAM).

[0164] To quantitatively elucidate the morphology of the fabricated NPSi, the NPSi material produced via the 60-min modified MACE process was analyzed at multiple length scales to obtain information on pore size and interpore spacing through statistical analysis. Figure 4Aand 4C Depicted are SEM images of NPSi materials produced via a modified MACE process using drop-cast 5 nm and 10 nm SiO2-AuNPs for 60 min, respectively (scale bar 500 nm). Figure 4B and 4D Shown respectively for 5nm and 10nm SiO2-AuNP Au core (particle) size distribution and the histogram of pore size distribution.Find that the average gold core diameter of commercial raw material 5nm SiO2-AuNP particle is 4nm+ / -1nm, and after MACE, the NPSi produced by the improved MACE method according to the present disclosure shows the gained average pore size of 6nm+ / -1nm and the porosity of 12.5%+ / -1% (across three different samples).The average diameter of commercial raw material 10nm SiO2-AuNP is 9nm+ / -2nm, and is shown as producing the average pore size of 8nm+ / -3nm and the porosity of 18.0%+ / -3% (across three different samples) in MACE process.Some differences between gold core size and final pore size can be attributed to the hole deformation that occurs due to SEM imaging, and this may be the result that high voltage electron beam affects Si.In addition, characterize gold core size and pore size by TEM and SEM respectively, this can also contribute to observing the inconsistency for 10nm sample. This approach is necessary to overcome the extreme difficulty of TEM sample preparation for high-porosity nanoporous silicon after MACE. Data from over 1000 pores and 400 particles were analyzed. Similar results were observed within a few microns of the substrate, and the consistency of etching and coverage was limited only by the quality and coverage of the SiO2-AuNP monolayer. Even if the particles did not assemble into a perfect monolayer, the SiO2 spacer ensured that the particles would be uniformly distributed. The minimum spacing, where d t is the particle diameter including the gold core and silica shell, and d AuNP is the gold core diameter for up to three layers of SiO2-AuNPs. However, the deposition of more than a monolayer results in a less controllable process due to the unpredictable reduction in interpore spacing and increase in porosity. Therefore, the ability to obtain large-area SiO2-AuNP monolayer arrays on Si surfaces is crucial for successful results. The sub-10 nm pore size and >12% porosity achieved here both represent an advance over the previous pioneering work of Gaborski et al., which showed NPSi with 10 nm to 40 nm pore sizes and 1.44% porosity.

[0165] For the application of NPSi in membranes and thin films, in addition to pore size and interpore spacing, pore depth and aspect ratio are also essential metrics. This was studied by filling the resulting NPSi pores with aluminum oxide (Al2O3) using atomic layer deposition (ALD) after MACE treatment. ALD was used for two purposes: (1) to preserve the nanoporous structure and (2) to enhance the image contrast of the pores relative to the Si substrate during cross-section milling and SEM imaging, respectively. Cross-sections were milled from 60-minute MACE samples using focused ion beam milling (FIB) and then imaged by SEM. Figures 5A to 5B The 5 nm ( Figure 5A ) and 10nm( Figure 5B ) SiO2-AuNP (scale bar is 500nm) SEM cross-sectional image of NPSi produced by improved MACE process. Figure 5C As shown, cross sections within (point 1) and below (point 2) the porous region were analyzed using energy dispersive x-ray spectroscopy (EDS). The highly conformal nature of ALD growth and its ability to fill narrow, high aspect ratio structures allow for accurate representation of the NPSi cross section. EDS analysis was used to verify pore filling within the pore region, showing the presence of Al and O within the bright regions (e.g., point 1) and the absence of these elements in the dark regions (e.g., point 2). Figures 5B to 5C For 5nm ( Figure 5A ) and 10nm( Figure 5B ) NPSi produced by SiO2-AuNP, EDS analysis shows that the pore depth exceeds 500nm, which is consistent with the pore size data ( Figure 4B ) combined, the resulting 5 nm NPSi had an aspect ratio exceeding 100:1. Such aspect ratios had not been previously demonstrated in NPSi produced by MACE, with the best recent result being the fabrication of nanogrooves with an aspect ratio of 65:1. A possible explanation for this result is the close proximity and consistent spacing of the catalytic AuNPs, which leads to uniform hole injection across the substrate, resulting in excellent anisotropy and the creation of high aspect ratio pores. Further support for this is provided by the observation that the 10 nm SiO2-AuNP catalyst produced shallower nanopores than its 5 nm counterpart after the same etching time ( Figures 5A to 5B A closer look at the inter-pore spacing of the 5nm and 10nm samples reveals more tightly packed pores in the 5nm case. It's possible that the closer proximity of the gold particles allows for more uniform hole injection, thus allowing for a greater etching rate, resulting in deeper pores. To characterize the etching rate over time, cross-sections of NPSi etched with a 5nm SiO2-AuNP catalyst were imaged after 30 and 60 minutes. Figure 6AAs shown in FIG, after 30 minutes of etching, the pore depth reaches about 200 nm, while 60 minutes of etching produces pores with a depth of about 600 nm, as shown in FIG. Figure 6B As shown ( Figures 6A to 6B The scale bar in the figure is 500 nm. The slower rate during the first 30 minutes of etching is likely the result of the time required to consume the silica shell and initiate etching. Therefore, a more accurate etch rate is obtained by considering the last 30 minutes of etching, during which pores are generated perpendicular to the surface at a rate of approximately 13 nm / minute. Due to the diffusion-limited nature of the etching process, this rate is not expected to be constant with increasing etching time.

[0166] The final important characteristic that contributes to the advantages of the produced NPSi is its chemical and physical stability. The demonstrated ability to deposit ultra-stable materials such as Al2O3 onto the high aspect ratio pore walls of the produced NPSi (Figure 5) shows that the material can be optimized for many different applications and to maximize chemical stability. The physical stability of the NPSi was studied by comparing images of freshly produced samples with samples stored in air for two months. In particular, Figures 7A to 7B SEM images of NPSi produced via a modified MACE process using drop-cast 5 nm SiO2-AuNPs and an etching time of 30 min are shown. Imaging was performed immediately after the etching process ( Figure 7A ), and imaging after a two-month period while the samples were in air ( Figure 7B ), scale bar is 500 nm. The morphological similarity between the two samples shows the structural stability of the material over time.

[0167] Example 2 - Precious Metal Island Catalyst

[0168] In another set of illustrative embodiments, an improved MACE process is used to produce NPSi with sub-10nm pore sizes and pore aspect ratios as high as 400:1. As discussed in more detail below, the method utilizes nucleation of noble metals sputtered on a silicon surface to form noble metal islands, which catalyze the etching process to form high aspect ratio pores. After etching, the porous structure is characterized by scanning electron microscopy (SEM) and transmission electron microscopy (TEM) as well as vertical or horizontal focused ion beam (FIB) cross-section milling at a depth of several microns within the silicon substrate. In addition, as described in more detail below, the NPSi is functionalized with Al2O3 and TiO2 via atomic layer deposition (ALD). The TiO2-functionalized NPSi exhibits a reflectivity of 6% to 8% for visible wavelengths and a reflectivity of 2% to 3% in the infrared, showing its promise as a robust and functional porous substrate. The developed method of MACE using a sputtered nucleation catalyst promotes the scalable manufacture of functional ultra-high aspect ratio nanopores in silicon.

[0169] Furthermore, the use of nucleated noble metal islands for the deposition of catalyzed nanopore etching via the MCE process offers the benefit of forming a uniform array of pores without the need to carefully control the parameters of the wet chemical process associated with monolayer formation, as may be required for nanoparticle catalysts. Instead, the deposited noble metals can naturally form highly ordered arrays of catalysts that are uniform in size and spacing. In this way, the method described herein can allow for a simple two-step MACE process for the production of NPSi. Furthermore, this method is capable of forming NPSi to form nanoporous structures over large areas without any inherent limitations on scaling.

[0170] (100) silicon wafers (B-doped, thickness: 275 ± 25 μm, resistivity: 0.001 Ω·cm to 0.01 Ω·cm) were solvent cleaned using a standard acetone-IPA-deionized water rinse. -8 Pa to 10×10 -8 The dried wafers were sputtered using an AJA ATC 2200 UHV sputter coater at a pre-deposition pressure of 1.5 Pa, an Ar flow rate of 40 sccm, and a deposition pressure of 4 mTorr. DC magnetron sputtering was performed on the target at 125 W. The deposition rate for gold (Au) was / second, for silver (Ag) After the noble metal is deposited and islands of the deposited noble metal are subsequently formed, the substrate is placed in a MACE solution (5.33 M HF, 0.12 M H2O2) to etch the substrate and form NPSi. The reaction is terminated by washing with water and removing the substrate, followed by drying with an N2 gun.

[0171] For the functionalized samples, NPSi was coated using a benchtop ALD system. In the case of the alumina functionalized samples, Al2O3 was coated via a static flow process at 200°C; the precursors were trimethylaluminum and H2O, and the calculated growth rate was / cycle. In the case of the titanium dioxide functionalized sample, TiO2 was coated via a static process at 190 °C; the precursors were tetrakis(dimethylamino)titanium and H2O, and the calculated growth rate was / cycle.

[0172] The process of sputtering a nominal metal film causes the nucleation of islands with uniform spacing and size. The size and morphology of the islands are affected by the surface defects on the substrate and the interface energy between the metal and Si, where the metal surface energy (γ) is γ Ag =1.246 J m -2 Up to 1.250 J m -2 , and γ Au =1.500 J m -2to 1.506 J m -2 Due to their wetting behavior on native oxide surfaces, Au and Ag films nucleate as isolated islands rather than continuous films.

[0173] Figures 8 to 9 show the sputtering SEM and TEM images of the nucleation islands generated by Au and Ag, and the corresponding histograms of island sizes. In particular, Figure 8A and 8B SEM and TEM images of sputtered nucleated Ag islands are shown, respectively, and Figure 8C A histogram of Ag island sizes is shown. Similarly, Figure 9A and 9B SEM and TEM images of Au islands are shown, respectively, and Figure 9C A histogram of Au island sizes is shown. The diameter of the observed nucleation islands is 4.4 ± 0.9 nm for Ag ( Figures 8A to 8C ), and 2.0±0.9nm for Au ( Figures 9A to 9C ). SEM images show large-area coverage of Au and Ag islands on the silicon substrate. Such island formation is uniform throughout the sputtered and etched area of ​​the silicon wafer. TEM images show Volmer-Weber film cores. The advantages of this method over conventional nanoparticle deposition or in-situ nanoparticle growth are excellent substrate-metal interface contact area, smaller particle size, narrower particle size distribution, and uniform catalyst spacing over a large area.

[0174] Figures 10A to 10B The histograms of the island spacing of Au and Ag samples are shown respectively. Figure 8B and 9B The spacing is calculated from the TEM images shown in . The spacing is defined as the separation distance between the center of one island and the center of its closest neighbor.

[0175] After sputter deposition, the wafer is placed in a MACE etchant solution. Silicon etching is achieved through localized silicon oxidation, facilitated by catalytic reduction of H2O2 on the noble metal nanoparticles. This is followed by HF etching of the SiO2, which continues etching of the metal islands perpendicular to the Si surface. In some cases, after etching, the nanopores of the NPSi are filled with Al2O3 via ALD to enhance imaging contrast and preserve the porous morphology prior to characterization.

[0176] In some cases, a further degree of control over island size can be employed via silicon surface treatment prior to catalyst nucleation. Removal of native oxides with HF prior to noble metal deposition increases the wettability of the film. This in turn is expected to result in lower metal island contact angles, but also a higher likelihood of silicide formation. The effect of contact angle on etching characteristics can be elucidated by comparing the etching behavior of Au and Ag metal islands, which exhibit different wetting behaviors. Ag, at its lower surface energy, has a larger contact angle with Si than Au, and thus forms more distinct islands, e.g. Figure 8A This leads to the <100> The etching in the direction is preceded by more lateral etching on the surface. This is observed as a linear track surrounding the hole, such as Figure 11B As shown, Figure 11B SEM image of a porous Si surface etched with Ag. In contrast, Au has a smaller contact angle and does not exhibit the same degree of lateral etching, as shown by Figure 11A It is confirmed that Figure 11A SEM image of the porous Si surface etched with Au. Figure 11C and 11D Figure 3 SEM images of focused ion beam (FIB) milled cross sections of Au and Ag etched NPSi samples at 1 hour etching time, respectively. The pores were backfilled with Al2O3 via ALD before FIB milling.

[0177] To verify the presence of nanopores and examine the morphology within the sample bulk after etching, horizontal cross-sections (ie, planar slices) of NPSi were compared with TEM images of control Si samples prepared via FIB at a depth of 4 μm. Figures 12A to 12B A schematic diagram of the process for preparing a planar sheet is shown. In particular, Figure 12A Shows the extraction of thin slices from NPSi samples via Omniprobe, Figure 12B TEM observation viewing angles are shown. Figure 12C TEM images of thin slices of native Si samples that were not exposed to the MACE process are shown. Figure 12D Shown is a TEM image of a thin slice of a sample sputtered with a gold catalyst and immersed in a MACE solution for 1 hour. Figure 12C and 12D The darker contrast area in the upper right corner is due to the platinum protective layer deposited before the planar sectioning. Figure 12D The TEM images shown in show the presence of pores at depths exceeding 4 μm within the substrate for the sample etched with sputtered Au. The ability to extract 100 nm flakes at a depth of 4 mm while maintaining the porous morphology also demonstrates the structural stability of the porous structure.

[0178] In addition to depositing Al2O3, this paper utilizes ALD to conformally coat the pore walls of NPSi with TiO2. This process creates a porous, high-surface-area functional substrate, whose porosity results in antireflective properties. Nanostructured TiO2 is being explored for applications due to its antireflective and self-cleaning properties. Figure 13 Reflectivity measurements of TiO2-functionalized NPSi are presented, showing reflectivity values ​​ranging from 2% to 8% in the UV-visible and short IR wavelengths for Au-etched Si and Ag-etched Si. The NPSi produced here shows promise due to its small pore size and high aspect ratio; using these metrics, a high pore density can be achieved, which correlates with a higher functional surface area. Therefore, these results demonstrate the potential of NPSi functionalized with deposited nanometer-thick TiO2 layers as a material with competitive light absorption and conversion properties.

[0179] Now turn Figure 14A , shows an SEM image of a FIB-milled cross-section of a TiO2-functionalized NPSi sample etched with Au. The TiO2-filled pore regions appear brighter compared to the silicon background. The white line shows the lowest visible pore depth along the cross-section. Figure 14B Shown Figure 14A The two differences shown in Figure 14A EDS analysis of the SEM areas (labeled 1 and 2 in FIG). Point 1 is at a depth of 700 nm below the surface and shows the presence of TiO2, while point 2 is at a depth of 1300 nm below the surface and does not show TiO2. The Si peak is cut off due to the strong signal from the underlying substrate. These results confirm that conformal TiO2 deposition within the high aspect ratio pores reaches a depth of approximately 1 μm below the substrate surface. Both Ti and O are present above the dashed line (700 nm, EDS point 1) ( Figure 12B ). In order to characterize, Figure 14A The pores shown in Figure 2 are completely filled with TiO2, but can also be partially filled to maintain the nanoporous structure of the substrate. EDS point 2 at 1300 nm shows the absence of Ti. C and Ga are present due to substrate and surface contamination and ion implantation, respectively. The absence of Ti signal indicates the absence of porous areas.

[0180] The coated NPSi surface was characterized via XPS to determine the composition of the ALD coating. Figure 15 High-resolution XPS scans of O and Ti from ALD-treated NPSi are shown. The O1 signal was deconvoluted to separate peaks associated with the SiO2 and TiO2 oxygen signals. These high-resolution elemental scans confirmed the stoichiometry of the TiO2 coating: deconvoluting the O1s peak into TiO2 and SiO2 yielded an O:Ti ratio of 2.3.

[0181] These results show that NPSi with the hole of sub-10nm, ultra-high aspect ratio is realized, and it can obtain functionality via TiO ALD deposition.By utilizing interface effect and film-forming property, sub-5nm noble metal island nucleates uniformly on multiple square centimeters on silicon.When compared with other MACE schemes, the technology developed is favourable, because the narrow size distribution of island and with the planar interface of Si reduce the generation of the undesirable etching effect caused by nanoparticle facet and substrate contact surface area.Shown for Au and Ag catalyst both, these islands etch out the hole of aspect ratio up to 400:1.Then use TiO via ALD functional layer coating the highly porous substrate of gained to show the potential functionality of NPSi.

[0182] Example 3 - Patterning of Two-Dimensional Materials

[0183] In another set of illustrative embodiments, nanoporous silicon (NPSi) films are used as mask materials for patterning MoS2 via oxygen plasma etching. This embodiment utilizes the methods herein, which allow the production of NPSi with sufficiently small (e.g., sub-30nm) holes with an aspect ratio greater than 10:1 in silicon, so that the holes penetrate a sufficiently thick free-standing film to provide mechanical stability over a large area. By utilizing the methods described herein to produce porous silicon films comprising nanopores with an aspect ratio greater than 1000:1 and a diameter less than 20nm, this embodiment shows the direct applicability of NPSi as an etching mask for the patterning of MoS2 and general 2D materials over any large area. In particular, this embodiment shows the patterning of regions with pores, which is achieved by using a silicon mask in which the nanoporous region is confined to the micrometer scale. After generating the nanoporous 2D monolayer, the pore size is further controlled by performing a mild thermal annealing in air, which is monitored by optical spectroscopy and electron microscopy.

[0184] A robust large-area NPSi film mask was fabricated by first sputtering a A silver film of nominal thickness was deposited, causing the nucleation of hemispherical nanoislands. The samples were then immersed in a solution of hydrogen peroxide and hydrofluoric acid (HF) for varying amounts of time to promote the etching of nanopores via a metal-assisted chemical etching (MACE) process, whereby oxidation occurs locally at the silicon-catalyst interface, followed by consumption of the oxide by HF. This electroless etching process is limited only by the presence of reactants in the solution and, when performed over extended periods of time, can produce pores with aspect ratios exceeding 1000:1 that completely penetrate Si substrates with thicknesses of many microns.

[0185] Two types of masks were explored in this example. The first type of mask is Figures 16A to 16CThe 50 nm thick (100) NPSi layer shown in FIG was etched for 1 to 2 minutes after catalyst deposition. In particular, Figure 16A An optical microscope image of five crystalline Si (c-Si) windows (100 μm by 100 μm) of 50 nm thickness on a single silicon mask chip (3 mm by 3 mm) is shown. The patterned mask defines the areas that are porous after etching. Figure 16A The inset in FIG shows a photograph of the entire silicon mask chip including ten 50 nm thick c-Si windows. Figure 16B TEM image showing a 50 nm thick c-Si window coated with Ag nano-island catalyst via radio frequency (RF) sputtering, and Figure 16C Shown is a TEM image of a porous 50 nm thick c-Si window after 1 minute of MACE. Figure 16C The inset in FIG shows sub-10 nm pores in the NPSi layer etched via the MACE process.

[0186] The second type of mask is a (100)NPSi layer of about 15 μm thick etched for 24 hours and is shown in Figures 16D to 16F In particular, Figure 16D Showing a thickness of 15 μm about 1 cm 2 Photograph of the NPSi mask. Figure 16E Show Figure 16D Figure 2. SEM cross-sectional image of a 15 μm thick NPSi mask. Due to the lack of fully anisotropic etching in the MACE process, a decrease in porosity with depth was observed. Figure 16F Show Figure 16E Figure 3 SEM image of the bottom surface of the NPSi mask (which is the surface in contact with the MoS2 flakes during oxygen plasma exposure).

[0187] Although the 50 nm thick NPSi mask makes the pore size less than 10 nm after 1 min etching ( Figure 16C ) or less than about 15 nm after 2 minutes of etching, but it is brittle and therefore only usable as a free-standing film in a 100 μm by 100 μm window. In contrast, a 15 μm thick mask can be used over many cm 2 The thick mask was produced on an area of ​​100 nm, but with a larger average pore size of 13.5 nm as indicated by Brunauer–Emmett–Teller (BET) nitrogen desorption analysis. The increase in pore size as measured by BET in the thick mask can be attributed to the highly roughened surface and the highly porous material ( Figure 16E ), although imaging of the smoother back surface in contact with MoS2 during patterning ( Figure 16F) also showed pore sizes larger than those present in thin masks. For a 50 nm thick mask, transmission electron microscopy (TEM) imaging of thousands of pores showed a porosity of approximately 2.8% ( Figure 16C ).

[0188] The schematic diagram of the patterning process of this embodiment is shown in Figure 17A middle. Specifically, Figure 17A The NPSi-assisted patterning process for nanoporous monolayer materials and the subsequent thermal annealing process for controlling porosity are shown. MoS domains (10 μm to 100 μm) grown by chemical vapor deposition (CVD) on a silicon wafer with a 300 nm silicon dioxide layer are spin-coated with poly(methyl methacrylate) (PMMA) and transferred to an NPSi substrate via KOH-assisted etching of the silicon dioxide layer. The substrate is selectively exposed to the O plasma on the opposite side of the MoS / PMMA to etch only the nanoscale partial regions of the MoS that are in contact with the nanopore regions of the NPSi mask. Subsequently, the NPSi mask is removed by KOH-assisted etching, and the MoS / PMMA layer is transferred to another silicon wafer. After removing the PMMA coating with acetone, the nanoporous MoS domains are examined by optical microscopy, scanning electron microscopy (SEM), photoluminescence (PL), and Raman spectroscopy. The initial porous structure is subsequently further modified by thermal annealing in air, and the structural changes are studied by the above-mentioned microscopy and spectroscopy techniques.

[0189] Imaging of MoS2 domains after oxygen plasma exposure through an NPSi mask showed the introduction of significant porosity into the material. Figure 17B and 17C The original MoS2 domains shown in Figures 17D to 17G The comparison between MoS2 domains patterned by selective etching through a thin (50 nm) NPSi mask is illustrated in FIG. In particular, Figure 17B SEM images showing pristine MoS2 domains on Si / SiO2 substrate, and Figure 17C Shown is an SEM image of a single triangular MoS2 domain that underwent the patterning process without exposure to O2 plasma due to being covered by a silicon mask chip. Figure 17D SEM images showing nanoporous MoS2 domains produced by selective etching through a thin (50 nm) NPSi mask for 2 minutes, Figure 17E Shown in Figure 17D SEM image of stellate nanoporous MoS2 domains found on the same substrate shown in. Figure 17F Depicts Figure 17E Magnified view of the area indicated by the yellow box, showing nanoscale pores with different diameters, and Figure 17GHigh magnification SEM image of nanoporous MoS2 domains is shown. MoS2 domains patterned with a thin NPSi mask ( Figures 17D to 17G ) has an average porosity of 8.5%, with a pore size of 65.9±13.3 nm.

[0190] Now refer to Figure 17H , depicts a WS2 domain patterned with a 15 μm thick NPSi mask (1 cm by 1 cm) on a 4 μm to 5 μm continuous area, specifically, Figure 17H Is to use 15μm thickness 1cm 2 SEM image of the nanoporous WS2 domains after 30 s of O2 plasma etching of the NPSi mask. Figure 17I yes Figure 17H A magnified view of the sample shown in , which shows the less round and connected morphology of the nanopores on the WS2 domains, and Figure 17J yes Figure 17H Another magnified view of the sample, showing Figure 17G These magnified images show that the morphology of the nanoporous domain is similar to that of the 15 μm thick NPSi mask. The density and size of the pores produced by the large-scale free-standing mask are similar to those of the Figures 17D to 17G This demonstrates the scalability of this patterning approach over a range of Si mask thicknesses and lateral sizes. Given the successful patterning of WS2 domains, it is expected that the application of the etching technique can be extended to a general set of 2D materials, including graphene, h-BN, and other transition metal dichalcogenides (TMDs).

[0191] In this example, O plasma etching through an NPSi mask completes in different timescales depending on the thickness of the mask. A 2-minute plasma treatment is used to pattern TMD domains through a 50 nm thick Si mask with sub-10 nm holes, while a 30-second exposure to plasma through a 15 μm thick Si mask is sufficient to produce holes of similar size in the domains.

[0192] Furthermore, after an initial 2-minute O₂ plasma treatment, which produces circular pores as determined by the shape of the NPSi structure, an effective strategy for controlling the pore size of monolayer MoS₂ domains was demonstrated via thermal treatment at 300°C in air. Defect sites in MoS₂ flakes tend to oxidatively etch under high-temperature conditions, making it possible to control the size of the nanopores via gradual thermal annealing. Nanopore size and density increased with heating time, indicating that mass loss occurs from the edges of existing nanopores, most likely in the form of sulfur loss. The enlargement of these pores leads to an increase in the edge-to-area ratio in the MoS₂ sample. This increase in edge sites, which has implications for optical and catalytic properties, can be effectively modified through a simple heating process without the need for a furnace or inert conditions. Depending on the application, an optimal edge-to-area ratio can be selected, enabling the realization of diverse nanoscale patterns.

[0193] Up to 30 minutes of annealing, oxidation did not produce any visible changes in the morphology of the nanopatterns under SEM. The effects of annealing for 30 minutes, 60 minutes, and 100 minutes are shown in Figures 18A to 18H , which shows the enlargement of pores and their convergence due to oxidation of edge sites of patterned MoS2. Figure 18A Shown is an SEM image of MoS2 domains after an initial 2 min O2 plasma treatment through a 50 nm thick NPSi mask. Figure 18B Show Figure 18A SEM image of MoS2 domains after thermal annealing in air for 30 min. Figure 18C shows an SEM image of MoS2 domains from the same sample after annealing for 60 minutes, and Figure 18D Shown is an SEM image of MoS2 domains from the same sample after annealing for 100 minutes. Figure 18E Show Figure 18A A magnified view of the sample shows individual pores with a diameter of approximately 70 nm. Figure 18F Show Figure 18B A magnified image of the sample shows the nanopattern of enlarged pores generated in the monolayer MoS2, and Figure 18G Show Figure 18C A magnified view of the sample shows further expansion and converging pores. Figure 18H Show Figure 18D A magnified image of a sample showing MoS2 flakes and oxidized MoS2 particles. The oxidized product (MoO3 particles) is shown in the magnified image as Figure 18H The bright spots are visible.

[0194] Figure 18I The effects of thermal annealing after plasma treatment on the edge density and porosity of MoS2 domains are summarized. Specifically, Figure 18Iis a plot of MoS2 edge density and porosity versus thermal annealing time for samples etched with O2 plasma. Figure 18I All data in , N = 3. At 300 °C, the porosity increases monotonically with annealing time, gradually increasing within the first 30 minutes and then increasing more sharply to reach 87.3 ± 2.61% after 100 minutes. The effect of annealing time on edge density is also depicted. It is observed that the edge density remains statistically similar after an annealing time of 30 minutes. This can be explained as follows: as the pores expand, the pores converge, resulting in a decrease in the edge to area ratio within the porous MoS2 domain. This equilibrium effect is observed to disappear after 60 minutes of annealing, mainly due to a sharp increase in pore intercalation, corresponding to a significantly larger porosity of 55.2 ± 10.1%, resulting in an edge density of 0.00827 ± 0.00170 nm after 30 minutes. -1 Nearly doubled to 0.0145±0.00106nm -1 Further annealing was observed to have a negative effect on the edge density, with the porosity increasing to 87.3 ± 2.61% after 100 min corresponding to a decrease in the edge density to 0.0107 ± 0.00135 nm. -1 , as the connectivity between the remaining MoS2 domains weakens. Our results provide valuable insights into optimizing edge density in MoS2 and other 2D materials via nanopatterning and thermal annealing.

[0195] Figure 18J The effectiveness of the developed process is shown, where half of the MoS2 domains placed on the 50 nm thick NPSi porous mask area are almost effectively patterned after O2 plasma exposure, while the other half located on the non-porous area remains pristine. Figure 18J An SEM image shows a MoS2 domain half on a 50nm-thick NPSi etch mask and half on non-porous silicon after O2 plasma treatment and 60 minutes of thermal annealing. This demonstrates seamless attachment of the MoS2 domain across the border of the 50nm-thick NPSi window on the mask to the supporting silicon chip, achieved via PMMA-assisted transfer and a mild drying process at 100°C. An SEM image showing a half-patterned MoS2 domain after 60 minutes of thermal annealing also confirms that under the same annealing conditions, oxidation occurs more extensively on the nanoporous side than on the pristine region. Figure 18Kis an SEM image of a pristine MoS2 domain after 100 minutes of thermal annealing in air, and depicts the formation of nanopores with diameters less than 50 nm on the pristine MoS2 domain after 100 minutes of thermal annealing in air. These pores, formed at a lower density than the O2 plasma-etched pores, are expected to grow on the basal plane of the MoS2 via preferential oxidation of defect sites and to be more severe at the domain edges and around existing cracks.

[0196] Understanding the gradual morphology changes induced by thermal annealing provides valuable insights into optimizing porosity, which is crucial for developing porosity-dependent applications such as nanofiltration and edge-site-specific applications including sensing and electrocatalysis.

[0197] like Figures 19A to 19F As shown, the visual changes in the domain morphology confirm the changes in optical properties measured by PL and Raman spectroscopy on pristine MoS2 and nanoporous MoS2. The optical properties of the MoS2 samples were first measured by PL spectroscopy at selected spots of approximately 1 μm size within the MoS2 domains. Figure 19A Figure 3 shows the normalized photoluminescence (PL) spectrum of the original MoS2 domain after thermal annealing, and shows that after excitation at 514.5nm, the original MoS2 domain before thermal annealing exhibits a PL spectrum with a peak at 683nm. When thermally annealed at 300°C, the PL increases and the peak blue-shifts to 680nm after 40 minutes of heating, to 675nm after 60 minutes of heating, and to 660nm after 100 minutes of heating. Previous work on block copolymer (BCP)-assisted nanopatterned MoS2 domains showed that the edge density increased and caused PL quenching, which is different from the observations in this embodiment. This is due to the difference in edge density formation, where nanodots, nanorods, and nanomeshes were not subjected to oxidation at high temperatures in previous work and were shown to have non-reconstructed edges by ADF-STEM.

[0198] like Figures 19B to 19C As shown in Figure 5, a more significant PL increase and blue shift are observed in the case of the nanoporous MoS2 sample. Specifically, Figure 19B shows the normalized PL spectra of nanoporous MoS2 domains after thermal annealing for up to 70 minutes, and Figure 19CShown are normalized PL spectra of nanoporous MoS2 domains after continued thermal annealing. O2 plasma-etched domains exhibit a PL spectrum with a peak at 677 nm, which blue-shifts to 664 nm after thermal annealing at 300°C for 20 minutes. Following the peak shift, the PL intensity increases when the domains are heated for up to 70 minutes, before further annealing leads to quenching of the PL due to severe degradation of the material. The PL variations seen in the nanoporous MoS2 samples follow the trend of the edge / area ratio, as determined by analysis of SEM images.

[0199] Laser-induced chemisorption oxidation of CVD-grown pristine MoS2 has been reported to cause a PL increase due to reduced n-doping, followed by a PL decrease after continued oxidation, and heating of exfoliated MoS2 in air also induces a rapid PL enhancement. Another report suggests that defect regions or cracks in a MoS2 layer exhibit a large PL enhancement compared to the pristine region after thermal annealing, due to heavy p-doping and reduced non-radiative recombination around the defective sites. A similar effect is seen in the nanoporous MoS2 sample shown in Figure 19. Compared to pristine MoS2, the sample etched with an NPSi mask exhibits a faster PL increase, which is related to the higher edge density and defects in the nanoporous domains. For the unetched MoS2 sample, only a slight PL increase is observed within the same high-temperature heating time range.

[0200] The contributions of exciton and trion emission to the PL spectrum were decoupled by multi-peak fitting to explain the overall blue shift and PL variation with thermal annealing time. Figure 17A) For the original MoS2 that was not etched but transferred multiple times, the PL was mainly composed of strong trion emission at about 1.83eV, and continuous thermal annealing induced exciton emission at about 1.87eV. After 100 minutes of annealing, the exciton / trion ratio was calculated to be 0.77. Nanoporous MoS2 exhibited similar behavior because after thermal annealing, the exciton contribution increased and dominated. Both the trion emission observed at about 1.83eV and the exciton at about 1.87eV grew rapidly after thermal oxidation, and the exciton / trion ratio increased from 0.53 after 20 minutes of annealing to 1.07 after 50 minutes of annealing, 1.19 after 60 minutes of annealing, 1.32 after 70 minutes of annealing, and 1.43 after 80 minutes of annealing. The exciton / trion ratio then dropped to 0.77 after 100 minutes of thermal treatment. This analysis shows that both the nanoporous and non-porous samples are highly n-doped before thermal annealing, as indicated by the initially dominant trion contribution, due to the PMMA transfer and O plasma etching processes. Both samples have similar levels of surface residues, and a comparison of the PL of both the etched and protected areas before high-temperature treatment shows similar spectra, indicating similar doping and defect densities in both cases.

[0201] like Figures 19D to 19E As shown, the Raman spectra of the two samples show an unchanged A 1g Pattern and Dropped E 1 2g model. Figure 19D shows the normalized Raman spectra of the pristine MoS2 domains after thermal annealing, and Figure 19E Shown is the normalized Raman spectrum of nanoporous MoS2 domains after thermal annealing. 1g / E 1 2g The increase in the ratio indicates that the doping level of the single-layer MoS2 flakes changes after thermal annealing, which is consistent with the change in the PL spectrum. 1 2g The negligible weakening of the mode shows that the strain variation within the MoS2 monolayer is not significant. Analysis of the time dependence of the heating process shows that an initial stage of minimum PL increase is observed, which may be due to the cleaning of surface residues, followed by a rapid onset of PL increase at 30 to 40 minutes (porous samples). The time scale of the rapid onset of PL is similar to previous studies on exfoliated MoS2 materials, where about 25 minutes at 300°C caused the PL maximum to be reached, followed by a rapid PL quenching during the next 25 minutes. The PL increase of the original sample is much slower than that of the nanoporous sample, which is related to the larger surface area of ​​the original MoS2 covered with the amorphous carbon protective layer and the fewer available edge sites prone to oxidation processes. The PL spectra are normalized to the Raman A1g peak, which normalizes the effects of different sample areas and material amounts. Figure 19F The relative maximum PL intensity is shown as a function of annealing time for nanopatterned MoS2 domains and pristine MoS2 domains.

[0202] The selective etching of the monolayer by the NPSi mask was determined via control experiments in which the CVD-grown pristine MoS2 domains were directly exposed to the O2 plasma for short (1 second, 10 seconds, and 30 seconds) times. The completely different results of the direct plasma etching can rule out the possibility that the NPSi is a layer that reduces the concentration and penetration of the O2 plasma. After 1 second exposure, the PL decreases and blue-shifts from 678 nm to 663 nm, as shown in Figure 2. Figure 20A (which shows the normalized PL spectra of the original MoS2 domains before and after 1 s direct O2 plasma treatment). This is not observed during the plasma etching through the NPSi mask. Figure 20B The Raman spectra in Figure 1 (which shows the normalized Raman spectra of pristine MoS2 domains before and after 1 second direct O2 plasma treatment) allow monitoring the effect of p-doping on pristine MoS2 after short plasma treatments. Any treatment longer than 1 second destroys the material, resulting in the absence of PL or Raman signals from the sample.

[0203] like Figures 20C to 20H As shown, the morphology changes after direct plasma treatment were studied by SEM. Figures 20C to 20E SEM images of pristine MoS2 domains after 1 second of direct O2 plasma treatment are shown at different magnifications. After 1 second of exposure to O2 plasma, the domains exhibit uniform cracking across the material without significant changes in the nanoscale morphology such as pore formation or defect enlargement. Figures 20F to 20H SEM images of pristine MoS2 domains after 10 seconds of direct O2 plasma treatment are shown at different magnifications. Longer treatments produce regions with shallower contrast and bright, needle-like material with lengths ranging from tens to hundreds of nanometers. The conversion of MoS2 to oxidation products is responsible for the disappearance of PL and Raman signals. This control experiment demonstrates the unique ability of NPSi masks to selectively form nanopores in 2D materials that exhibit bright PL even after nanopatterning and severe etching.

[0204] In the above examples, the NPSi mask was made from an initially 300 μm thick (100) DSP wafer (resistivity 1 Ω·cm to 10 Ω·cm, p-type (boron doped)) thinned at 70°C via a 50 nm thick (100) crystalline Si film (SiMPore Inc.) and a 15 μm thick (100) crystalline Si film. Both types of substrates were cleaned in acetone, isopropanol and deionized water and dried under nitrogen before catalyst deposition. Hemispherical nanocatalysts were deposited via RF magnetron sputtering using an ATC6 target sputtering tool (AJA International). In all experiments, Ag was deposited onto the substrate at an RF power of 30 W. Immediately after catalyst deposition, the sample was immersed in a MACE solution containing 5.33 M HF and 0.12 M H2O2 for 1 minute in the case of a 50 nm thick mask, and for 24 hours in the case of a 15 μm thick mask. The mask was thoroughly rinsed with deionized water and dried under nitrogen after etching.

[0205] The monolayer MoS with CVD growth is spin-coated on the substrate at 500 rpm for 5 seconds, 2000 rpm for 10 seconds, and 4500 rpm for 45 seconds. The sample is annealed at 180 ℃ for 90 seconds to evaporate anisole, and by dissolving the substrate in 1M KOH solution overnight to remove it. The PMMA layer in the MoS domain is then rinsed with deionized water and transferred to the NPSi substrate. The sample is dried at 150 ℃ for 10 minutes and turned upside down to be selectively exposed to oxygen plasma. The oxygen plasma etching of MoS is carried out by using Harrick Scientific PDC-32G plasma cleaner (18W, oxygen flow rate is 0.6 SCFH (standard cubic feet / hour)) through the NPSi mask. The porous MoS formed on the PMMA layer is then transferred to new Si / SiO substrate via the KOH etching of the NPSi as described in the initial transfer process.

[0206] Use equipped with Ar + PL and Raman spectra were acquired on a Horiba LabRAM800HR spectrometer equipped with a 514.5 nm excitation source and a Peltier-cooled CCD detector. The laser was focused on the sample using a 100x objective through a 400 nm confocal aperture under reflected illumination. The laser spot on the sample had a diameter of approximately 1 μm and had a power of approximately 4 mW at the sample surface. Scanning electron microscopy was performed using a Zeiss Ultra Plus FESEM.

[0207] Example 4 - Nanofiltration membrane

[0208] In another set of illustrative embodiments, inorganic nanofiltration (NF) membrane materials formed of nanoporous silicon (NPSi) were studied. NPSi membranes are capable of performing separations at the sub-1 nm molecular scale while withstanding a variety of chemically and thermally harsh environments. The membranes are produced in three steps that facilitate large-scale, economically viable manufacturing. This example introduces an in-depth description of the manufacturing process, characterization of the resulting membrane materials, and an evaluation of the filtration performance of four NPSi membranes using industrially relevant feed streams, including a demonstration of zero liquid discharge (ZLD) performance. Taken together, these results show that NF membranes have superior scalability, cost, and rejection performance relative to ceramic membranes at the sub-1 nm scale.

[0209] NPSi membrane via Figure 21A The method involves chemical thinning of commercial silicon, deposition of a metal catalyst, and top-down chemical etching of ultrahigh aspect ratio nanopores through the thinned Si layer. For all samples, a 5 cm crystalline Si wafer with a (100) orientation was selectively exposed to a hot potassium hydroxide (KOH) bath for several hours to produce a thin Si foil of 30 nm diameter across the center of the wafer. A nominal sub-nm silver layer was then deposited by sputtering on the unetched opposite side of the wafer at low power for a very short exposure time. Finally, the wafer was inserted into a chemical bath containing hydrogen peroxide (H2O2) and hydrofluoric acid (HF) to promote the selective oxidation and removal of Si at the Ag-Si interface by a metal assisted chemical etching (MACE) process. The result is a monolithic silicon membrane with an external treatment layer of the thickness of the initial wafer. The specific manufacturing conditions for each of the four membranes discussed in this example are shown in Table 1 below. The planar surface of the wafer where MACE occurred exhibits a beige color due to its microscale roughness, while the opposite surface exhibits a beige color due to its microscale roughness. Figures 21B to 21D The MACE shown in FIG appears matte gray due to nanoscale roughness caused by KOH etching and penetration of Ag catalyst particles into the surface. Specifically, Figure 21B Photographs showing the front and back surfaces of the original film, Figure 21C SEM showing the initial etched surface, and Figure 21D SEM images of the surface opposite to the initial etched surface are shown. Figure 21E The SEM image of the membrane cross section shown shows that the porosity decreases with increasing depth through the Si, resulting in an asymmetric structure. All scale bars are 2 μm, and Figure 21DThe scale bar in the inset is 200 nm. Fabrication was performed in such a way that the smoother side of the membrane was exposed to the feed stream during filtration, as reducing the surface roughness is understood as a method for minimizing membrane fouling. The BJH pore size distribution calculated from BET nitrogen adsorption and desorption experiments is shown in Figure 21F and are discussed throughout this Example as they relate to filtration performance.

[0210] Table 1

[0211]

[0212] Prior to filtration testing, the NPSi membranes were characterized for thickness, surface hydrophilicity, and pore size distribution. A static surface contact angle of 47.9° ± 2.1° was observed for the pristine membranes. In some samples, the membranes were treated with piranha solution immediately after MACE to aggressively oxidize their surfaces. After this treatment, the same membrane samples exhibited a reduced contact angle of 12.2° ± 1.7°. Increased hydrophilicity is generally considered an important metric for reducing membrane fouling rates. Figure 22A and 22B Images of static water contact angles on pristine and piranha-treated films are shown, respectively.

[0213] For the membrane produced by the standard method, the Brunauer-Emmett-Teller (BET) surface area was measured to be 6.20 m 2 / g and the Barrett-Joyner-Halenda (BJH) average pore size is 18.96nm, calculated using adsorption isotherms. The pore size distribution shows a clear peak near the pore size of 24nm, with a more complex curve for pores less than 10nm, including a monotonically increasing number of pores as the pore size decreases to below 3.6nm. The most likely explanation for this observation is that there are multiple types of porous morphologies within the membrane layer, the first being larger pores found within a few microns of the initial etched surface, and the second being dead-ends and through-holes deep within the bulk. The latter structure is often etched only by a single Ag catalyst particle and is less prone to pore expansion given its high aspect ratio and depth within the bulk membrane. These pores are therefore expected to have significantly smaller diameters, which are determined by, for example, Figure 21F This is indicated by the presence of a large number of pores smaller than 3.6 nm.

[0214] The filtration of 5nm diameter gold nanoparticles (AuNPs) suspended in aqueous solution was studied in a dead-end membrane configuration without stirring to characterize the rejection and permeability behavior of the NPSi membrane over time. The original membrane (M1) was inserted into the cell and then completely immersed in 5mL IPA at a pressure of 10.34 bar (150psi) to maximize pore wetting. 100mL of deionized water was then added to the cell and placed at 10.34 bar to evaluate pure water permeability.

[0215] After an initial stabilization period, the permeability was observed to be constant at 0.31 LMH / bar over a period of 30 hours, as shown in FIG. Figure 23 (which shows a plot of permeate mass versus time for M1 at constant pressure during pure water and 5 nm AuNP rejection experiments). Without drying the membrane, an aqueous suspension of 5 nm AuNPs was added to the cell to produce a feed concentration of 3.7e13 particles / mL, and the pressure was reapplied at 10.34 bar. Continuous monitoring of permeability and collection of approximately every 10 g of permeate sample revealed a trade-off between permeability and rejection, as shown in FIG. Figure 24A (which depicts a graph of permeability and retention for M1 during a 5 nm AuNP filtration experiment). Although the permeability decreased from the pure water value to 0.13 LMH / bar within the first three minutes of the experiment and eventually dropped to less than one-third of its initial value ( Figure 23 ), but the retention rate increased from the initial value of 84% and quickly reached a plateau above 94%. This was confirmed by: Figure 24B Photographs of samples collected during the experiment shown in , thus showing the entrapment of 5 nm Au particles and the concentrated retentate solution.

[0216] These findings indicate that the pristine membrane was able to retain a large fraction of the 5 nm particles, with the retention likely being primarily due to steric size exclusion effects, as evidenced by the BJH pore size distribution. As filtration continued, the decrease in flux coupled with the increase in retention suggested that the pores were becoming clogged either at the pore entrance or within the membrane bulk. SEM imaging of the membrane surface after the experiment indeed revealed less adsorption of AuNPs to the membrane surface ( Figure 24C ), which may preferably occur in regions where there is a higher flux of solutes into the pore opening. Specifically, Figure 24CFigure 2 is an SEM image of the surface of M1 after the test. Comparison of the initial feed and final retentate concentrations shows a particle concentration factor of 3.83, while the bulk concentration factor, ignoring evaporation, is 3.72. Although the permeate collection bottle was isolated from the surrounding environment, the lack of an airtight seal allowed evaporation to occur, meaning that the actual bulk concentration factor was greater than 3.72. However, the close correspondence between the particle and bulk concentration factors indicates that Au particles did not accumulate to a significant extent on the membrane surface or within the membrane bulk, despite the presence of concentration polarization due to the lack of agitation during filtration. Figure 24D The SEM cross-sectional image shown in Figure 3 shows a thickness of 34.3 μm and a low percentage of pores penetrating from the feed to the permeate side of the active layer. <110> The known tendency of lateral etching in the direction perpendicular to the Si surface <100> The low porosity on the feedstream side of the membrane, with similar frequencies in the direction, is not surprising. For the catalyst to create pores extending directly through the Si membrane, the pore aspect ratio would need to exceed 9500:1 for pores less than 3.6 nm, explaining the high probability of etching direction changes. Consequently, the estimated surface porosity on the feedstream side of the membrane is less than 1%, presumably the primary reason for the generally low permeability measurements observed.

[0217] To explore the ability of the NPSi membrane to retain charged solvated species smaller than 5 nm AuNPs, filtration experiments were performed using a feed stream consisting of aqueous molecular dye solutions of reactive black (RB) and methyl orange (MO), both negatively charged dyes with MWs of 992 g / mol and 327 g / mol, respectively. Such dye species not only serve as effective analogs for many industrially important small charged molecules, but are also directly relevant to the textile industry, a $870B market that increasingly requires the removal of dyes from effluents using nanofiltration. The dye experiments were conducted in the same manner as the AuNP experiments discussed above (with pure water followed by 10 -4 The membrane was tested in a similar manner to that of the membrane (M dye solution, passed through a dead-end structure at an applied pressure of 10.34 bar). In this case, the membrane was not pre-wetted with IPA. Two original membranes, M2 and M3, were tested, wherein each membrane was thoroughly rinsed with pure water between dye tests. Figures 25A to 25E The pure water permeability and RB and MO rejection for the two membranes are shown in FIG. In particular, Figure 25A a graph showing the pure water permeability and the rejection of the molecular dyes MO and RB for M2 and M3, and Figures 25B to 25D Shown for M2( Figure 25B and 25C ) and M3( Figure 25D and 25E )Filter RB( Figure 25B and 25D ) and MO( Figure 25C and25E ) photos of feed (left) and permeate (right) samples.

[0218] M2, produced using the same process as M1 but with a shorter etching time, exhibited a lower initial pure water permeability of 0.11 LMH / bar, likely due to the lack of pore wetting with IPA and general variability in the pore etching process. After the permeability test, M2 retained 91% of RB and 70% of MO ( Figures 25A to 25C ), with thorough rinsing with pure water between each test. A small drop in permeability was observed during each test. With the goal of obtaining a tighter MWCO, M3 was produced using slightly modified catalyst deposition parameters (4 seconds of sputtering time and 15 W of power versus 2 seconds and 30 W) with the aim of reducing the average catalyst and subsequent pore size. This membrane exhibited a reduced pure water permeability of 0.04 LMH / bar and an improvement in the rejection of the two dye species RB and MO to 99% and 92%, respectively ( Figure 25A 、 25D and 25E). This result shows the ability to retain molecules with a size of 327 g / mol due to both steric and charge-based interactions.

[0219] An important property of a filtration membrane is its ability to resist the accumulation of fouling and scaling on its surface, a property understood to be primarily affected by surface roughness and hydrophilicity. Oil-water separation, an important process in multiple industries (including oil and gas and food and beverage), has in many cases proven to be particularly difficult for polymer membranes due to the large amount of fouling. In order to evaluate the performance of NPSi membranes in oil-water separation applications with respect to antifouling and cleanability, a feed solution of 30 wt% hexadecane emulsified in water was introduced into a dead-end cell and fed through an original sample (M4) at 3.45 bar (50 psi). M4 was produced using the same conditions as the previous three samples, but the sputtering power was increased to 45 W with the goal of increasing porosity and subsequent permeability. For the original membrane before piranha treatment and without IPA pre-wetting, the resulting pure water permeability was 0.67 LMH / bar. The feed hexadecane emulsion was characterized by DLS, showing an average particle size of 389.4 ± 5.1 nm with a PDI of 0.257 ± 0.003 (Figure 6K). A series of experiments were conducted sequentially from pure water to similar emulsion feed streams at approximately 50% volume recovery, with different cleaning protocols employed with the goal of regenerating membrane flux after each experiment. In order, pure water, emulsion, emulsion after piranha treatment, emulsion after heat treatment at 250°C, and emulsion after IPA rinse were tested. A final experiment, representing ZLD operation, was conducted in which the maximum permeate volume was achieved by subjecting the feed emulsion to applied pressure until no further permeate could be extracted.

[0220] A series of experiments showed promising oil-water separation performance in combination with flux regeneration after hydrocarbon fouling. Figure 26A and 26B Shown are photographs of the feed, permeate, and concentrate for oil-water separation tests conducted using the M4. Figure 26C is a photograph showing the absorption of hexadecane from the surface of M4 after the experiment.

[0221] Thermogravimetric analysis of all feed and permeate samples showed similar permeate purity, while showing that the permeation rate in each run was greater than the initial emulsion run, reaching a peak after piranha treatment, likely due to increased surface hydrophilicity. Figure 26D Shown are TGA measurements of pure water and hexadecane superimposed on the feed and permeate. Figure 26E and 26F Optical microscope images of a 50-fold diluted feed and undiluted permeate are shown, respectively.

[0222] The final ZLD experiment was able to obtain a total mass recovery of 62.9% permeate from a feed solution consisting of 30% hexadecane. Assuming pure water permeation, as indicated by TGA, a mass recovery of 89.8% water was achieved in a dead-end configuration without stirring. Figures 26G to 26I A photograph showing solid hexadecane recovered during the ZLD test, and Figure 26J TGA measurements of the results from the ZLD test (including feed, permeate and concentrate) are shown, showing that effective separation of hydrocarbons from water was achieved. Figure 26K Shown are the droplet size distributions measured using DLS for an emulsified hexadecane emulsion feed solution.

[0223] Every cleaning regimen attempted, including simple rinses with IPA and water, proved effective in regenerating membrane flux, although the membrane was also shown to be extremely chemically and thermally resilient and to withstand the piranha solution.

[0224] Example 5 - Thick film etched from both sides

[0225] In yet another illustrative embodiment, another method for forming the film was investigated that may provide benefits in terms of improved commercial viability. Unlike the embodiments discussed above, in this embodiment, the silicon is not thinned prior to etching, and the catalyst is deposited directly on the opposite side of a thicker Si wafer (e.g., 200 μm to 400 μm thick), as shown in FIG. Figure 27 The cross-sectional SEM image is shown in FIG.

[0226] The catalyst deposition step is the same as in the previous examples (silver, 30W, 2 seconds), but in this example it is performed on each side of the wafer. In addition, the hole etching is performed in the same chemical solution as in the previous examples, but it is performed for a longer time (e.g., 24 hours to 48 hours) to allow significant etching to occur. Using these methods, the aspect ratio of the holes that completely penetrate the thicker membrane can be very high. For example, Figure 28 The pore size distribution is shown for a 400 μm thick film and shows a significant presence of pores with a diameter of about 6 nm. In order to penetrate the full thickness of the wafer, the aspect ratio of these pores would be greater than 65,000:1.

[0227] Although several embodiments and examples of the present invention have been described and illustrated herein, a person of ordinary skill in the art will readily appreciate a variety of other ways and / or structures for performing the functions described herein and / or obtaining the results and / or one or more advantages described herein, and each such variation and / or modification is considered to be within the scope of the present invention. More generally, a person skilled in the art will readily understand that all parameters, dimensions, materials, and configurations described herein are intended to be exemplary, and that the actual parameters, dimensions, materials, and / or configurations will depend on the specific application or applications in which the teachings of the present invention are used. A person skilled in the art will recognize or be able to determine, using only routine experimentation, many equivalents to the specific embodiments of the present invention described herein. Therefore, it should be understood that the foregoing embodiments are presented by way of example only, and within the scope of the appended claims and their equivalents, the present invention may be implemented in ways other than those specifically described and claimed. The present invention relates to each individual feature, system, article, material, kit, and / or method described herein. In addition, any combination of two or more such features, systems, articles, materials, kits, and / or methods is included within the scope of the present invention if such features, systems, articles, materials, kits, and / or methods are not mutually inconsistent.

[0228] Unless explicitly stated to the contrary, as used herein in the specification and claims, a noun without a quantifier should be understood to mean "at least one."

[0229] As used herein in the specification and claims, the phrase "and / or" should be understood to mean "one or both" of the elements so connected, i.e., in some cases the elements are present simultaneously, while in other cases the elements are present separately. Unless expressly indicated to the contrary, other elements may optionally be present in addition to those expressly indicated by the "and / or" conjunction, whether related or unrelated to those elements expressly indicated. Thus, as a non-limiting example, when used in conjunction with open language such as "comprising," a reference to "A and / or B" may, in one embodiment, refer to A without B (optionally including elements in addition to B); in another embodiment, to B without A (optionally including elements in addition to A); in yet another embodiment, to both A and B (optionally including other elements); and so on.

[0230] As used herein in the specification and claims, "or" should be understood to have the same meaning as "and / or" as defined above. For example, when separating the items in a list, "or" or "and / or" should be interpreted as inclusive, that is, including at least one of a plurality of elements or a list of elements, but also including more than one, and optionally including items not listed in addition. Only when the opposite terms are clearly indicated, such as "only one" or "exactly one", or "consisting of when used in the claims, refers to including exactly one element in a plurality of elements or a list of elements. Generally, the term "or" as used herein, when followed by an exclusive term (such as "either", "one of", "only one" or "exactly one"), should only be understood as an exclusive alternative (i.e., "one or the other, but not both"). When used in the claims, "consisting essentially of" should have its ordinary meaning used in the field of patent law.

[0231] As used herein in the specification and claims, when referring to a list of one or more elements, the phrase "at least one" should be understood to mean at least one element selected from any one or more elements in the list of elements, but does not necessarily include at least one of each and every element explicitly listed in the list of elements, and does not exclude any combination of elements in the list of elements. This definition also allows that elements other than the elements explicitly specified in the list of elements to which the phrase "at least one" refers may optionally be present, whether related or unrelated to those explicitly specified elements. Thus, as a non-limiting example, "at least one of A and B" (or equivalently, "at least one of A or B," or equivalently, "at least one of A and / or B") may refer to at least one A, optionally including more than one A, but no B (and optionally including elements other than B) in one embodiment; at least one B, optionally including more than one B, but no A (and optionally including elements other than A) in another embodiment; at least one A, optionally including more than one A, and at least one B, optionally including more than one B (and optionally including other elements) in yet another embodiment; and so on.

[0232] In the claims and throughout the foregoing description, all transitional phrases such as "comprises," "comprising," "with," "having," "containing," "involving," "having," and the like are to be construed as open-ended, meaning including, but not limited to, including. Only the transitional phrases "consisting of" and "consisting essentially of" shall be closed or semi-closed transitional phrases, respectively, as set forth in the United States Patent Office Manual of Patent Examining Procedures Section 2111.03.

[0233] Unless otherwise defined or indicated, any terms used herein that refer to, for example, the shape and / or geometric relationships of or between one or more articles, structures, and / or subassemblies thereof, and / or combinations thereof and / or any other tangible or intangible elements not listed above that are represented by such terms should be understood not to require absolute conformity with the mathematical definition of such terms, but rather to indicate, to the extent possible, the features that conform to the mathematical definition of such terms as those skilled in the art would understand to be most closely related to such subject matter. Examples of such terms related to shape and / or geometric relationships include, but are not limited to, terms describing shapes, such as round, square, circular / circular, rectangular / rectangular, triangular / triangular, cylindrical / cylindrical, elliptical / elliptical, (n)polygonal / (n)polygonal, etc.; surface and / or bulk material properties and / or spatial / temporal resolution and / or distribution, such as smooth, reflective, transparent, clear, opaque, rigid, impermeable, uniform, inert, non-wettable, insoluble, stable, unchanging, constant, uniformly, etc.; and many other terms that will be apparent to those skilled in the relevant art. As an example, a manufactured article described herein as a "square" does not require that such article have faces or sides that are completely planar or linear and intersect at exactly 90-degree angles (indeed, such an article can only exist as a mathematical abstraction), but rather, as will be understood by those skilled in the art, the shape of such an article should be interpreted as approximating a mathematically defined "square" to the extent that is generally achievable and achievable with the enumerated manufacturing techniques.

Claims

1. A porous semiconductor material comprising: semiconductor materials; and a plurality of pores in the semiconductor material, the plurality of pores having an average pore size of less than 20 nm, and wherein the plurality of pores defines a total volume porosity of at least 0.1% measured by dividing the total pore volume by the sum of the total pore volume plus the volume of the solid material, wherein at least 0.05% of said pores extend through said material from one surface to an opposite or different surface, and wherein the material has a thickness that is a smallest cross-section of the material, and the thickness is at least 0.05 microns.

2. The porous semiconductor material of claim 1, wherein the plurality of pores defines a total volume porosity of less than 50%.

3. The porous semiconductor material of claim 2, wherein the plurality of pores defines a total volume porosity of 1% to 30%.

4. The porous semiconductor material of claim 1, wherein less than 30% of the pores extend through the material from one surface to an opposite or different surface.

5. A porous semiconductor material according to claim 4, wherein 0.5% to 20% of the pores extend through the material from one surface to an opposite or different surface.

6. A porous semiconductor material according to claim 5, wherein 1% to 10% of the pores extend through the material from one surface to an opposite or different surface.

7. The porous semiconductor material of claim 1, wherein the material has a thickness of less than 400 microns.

8. The porous semiconductor material of claim 7, wherein the material has a thickness of at least 10 microns.

9. The porous semiconductor material of claim 8, wherein the material has a thickness of at least 100 microns.

10. The porous semiconductor material of claim 9, wherein the material has a thickness of at least 200 microns.

11. The porous semiconductor material of claim 10, wherein the material has a thickness of at least 250 microns.

12. The porous semiconductor material of claim 1, wherein the plurality of pores are etched into the semiconductor material. The porous semiconductor material according to claim 1 , wherein the average inter-pore spacing of the plurality of pores is less than 10 nm. The porous semiconductor material of claim 1 , wherein the average aspect ratio of the plurality of pores is greater than 100:

1.

15. The porous semiconductor material of claim 1, wherein the average aspect ratio of the plurality of pores is greater than 1000:1.

Images