Methods of depositing nanostructures on a substrate and nanostructure arrays

Abstract

A method of depositing nanostructures on a substrate includes forming a patterned alignment layer on a surface of the substrate, wherein the patterned alignment layer has one or more cavities, each cavity having a bulk region for containing at least one nanostructure template therein and a plurality of extension regions extending from and in fluid communication with the bulk region, wherein the plurality of extension regions are sized and shaped to not contain the at least one nanostructure template; and diffusing a nanostructure template into the one or more cavities of the patterned alignment layer.

Description

[0001] Cross-references to related applications

[0002] This application claims priority to PCT applications PCT / CN2020 / 082375, PCT / CN2020 / 082377, and PCT / CN2020 / 082778, all filed on March 31, 2020, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This application relates primarily to nanofabrication technology, and more specifically, to a method for depositing nanostructures on a substrate and an array of nanostructures fabricated by the method. Background Technology

[0004] Artificial nucleic acid nanostructures and other types of nanostructures can self-assemble into specified templates for spatially arranging functional materials at the deep submicron and nanoscale. Due to the controllable assembly of the templates, functional materials or nanostructures (e.g., carbon nanotubes (CNTs), metal nanorods or nanoparticles, polymer nanoparticles, and silicon nanowires) bound to these templates can be positioned regularly and precisely.

[0005] Various techniques have been developed to integrate nanostructure templates with conventional microfabrication techniques. For example, patterned alignment layers can be formed on a substrate to facilitate and guide the arrangement of individual nanostructure templates. However, the yield and controllability of this approach remain unsatisfactory, reducing its prospects for use in industrial applications.

[0006] Therefore, further improvements are needed to existing nanofabrication technologies. Summary of the Invention

[0007] The purpose of this application is to provide a method for depositing nanostructures on a substrate.

[0008] According to one aspect of this application, a method for depositing nanostructures on a substrate is provided, the method comprising: forming a patterned alignment layer on a surface of the substrate, wherein the patterned alignment layer has one or more cavities, each cavity having a main region for receiving at least one nanostructure template therein and a plurality of extension regions extending from and in fluid communication with the main region, wherein the size and shape of the plurality of extension regions are configured not to receive at least one nanostructure template; and diffusing the nanostructure template into the one or more cavities of the patterned alignment layer.

[0009] In some implementations, the method further includes removing the patterned alignment layer from the substrate.

[0010] In some implementations, removing the patterned alignment layer from the substrate includes using a stripping process to remove the patterned alignment layer from the substrate.

[0011] In some embodiments, the main body region of each of the one or more cavities is further configured to be oriented to house at least one nanostructure template therein.

[0012] In some embodiments, each of the one or more cavities is further configured such that when at least one nanostructure template is housed in the cavity, at least one translational degree of freedom of the at least one nanostructure template along the surface of the substrate is restricted.

[0013] In some embodiments, forming a patterned alignment layer on the surface of the substrate includes: forming an alignment layer on the surface of the substrate; and a patterned alignment layer.

[0014] In some implementations, the patterned alignment layer includes an alignment layer patterned using a photolithography process.

[0015] In some embodiments, the photolithography process includes electron beam lithography, photolithography, imprint lithography, or directional self-assembly of bulk copolymers.

[0016] In some embodiments, forming an alignment layer on the surface of the substrate includes spin-coating the surface of the substrate with photoresist to form the alignment layer.

[0017] In some embodiments, the photoresist includes polymethyl methacrylate (PMMA) or polydimethylsiloxane (PDMS).

[0018] In some embodiments, diffusing a nanostructure template into one or more cavities of a patterned alignment layer includes: immersing a solution containing the nanostructure template on the patterned alignment layer; and cultivating a culture substrate to diffuse the nanostructure template into the cavity.

[0019] In some embodiments, the culture substrate includes dehydrating or evaporating the substrate in a sealed chamber for a predetermined period of time.

[0020] In some embodiments, the method further includes: differentiating the substrate to improve the adhesion of the substrate surface to the nanostructure template before diffusing the nanostructure template into one or more cavities of the patterned alignment layer.

[0021] In some embodiments, the nanostructure template is modified with a functional nanostructure, wherein, after removing the patterned alignment layer from the substrate, the method further includes: removing the template portion of the nanostructure template from the substrate to leave the functional nanostructure on the substrate.

[0022] In some embodiments, the method further includes forming a fixation structure on the substrate to fix the functional nanostructure on the substrate before removing the template portion of the nanostructure template from the substrate.

[0023] In some embodiments, one of the cavities has multiple extension regions that increase the volume of the cavity by at least 5%, or preferably by 10%, 20%, 30%, 50%, 100%, 150%, 200%, or more.

[0024] In some embodiments, the depth of each cavity in one or more cavities is two or more times the thickness of the nanostructure template, preferably five or more times.

[0025] In some embodiments, the nanostructure template comprises one or more substances selected from the group consisting of: nucleic acid templates, modified nucleic acid templates, protein templates, polymer templates, peptide nucleic acid templates, carbon nanotubes (CNTs), polymer-encapsulated CNTs, CNT films, semiconductor nanoparticles, semiconductor nanowires, semiconductor nanoblocks, metal nanoparticles, metal nanowires, metal nanoblocks, polymeric nanoparticles, polymeric nanowires, polymeric nanoblocks, ceramic nanoparticles, ceramic nanowires, ceramic nanoblocks, metal oxide nanoparticles, metal oxide nanowires, metal oxide nanoblocks, fluoride nanoparticles, fluoride nanowires, and fluoride nanoblocks.

[0026] In some implementations, the nanostructure template is modified with a functional nanostructure.

[0027] In some embodiments, the functional nanostructure comprises one or more substances selected from the group consisting of: carbon nanotubes (CNTs), polymer-encapsulated CNTs, CNT films, semiconductor nanoparticles, semiconductor nanowires, semiconductor nanoblocks, metal nanoparticles, metal nanowires, metal nanoblocks, polymeric nanoparticles, polymeric nanowires, polymeric nanoblocks, ceramic nanoparticles, ceramic nanowires, ceramic nanoblocks, metal oxide nanoparticles, metal oxide nanowires, metal oxide nanoblocks, fluoride nanoparticles, fluoride nanowires, and fluoride nanoblocks.

[0028] According to another aspect of this application, a nanostructure array is provided, the nanostructure array comprising at least one nanostructure template deposited on a substrate by the methods of the foregoing aspects.

[0029] According to another aspect of this application, a nanostructure array is provided, the nanostructure array comprising at least one functional nanostructure deposited on a substrate by the methods described in the foregoing aspects.

[0030] The foregoing is an overview of this application, which may be simplified, summarized, and details omitted. Those skilled in the art will understand that this section is merely illustrative and not intended to limit the scope of this application in any way. This summary section is neither intended to identify key or essential features of the claimed subject matter nor to serve as an aid in determining the scope of the claimed subject matter. Attached Figure Description

[0031] The foregoing and other features of this application will be more fully understood through the following description taken in conjunction with the accompanying drawings and the appended claims. It should be understood that these drawings depict only some embodiments of the application and should not be construed as limiting the scope of the application. The content of this application will be shown more clearly and in more detail through the accompanying drawings.

[0032] Figure 1a illustrates the dynamic binding process of DNA origami on a patterned surface, while Figures 1b and 1c show the alignment of the DNA origami on the substrate after the binding process.

[0033] Figure 2 A method for depositing nanostructures on a substrate according to one embodiment of this application is shown.

[0034] Figure 3 , Figures 4a to 4b and Figures 5 to 7 An exemplary graphical alignment layer on a substrate is shown according to an embodiment of this application.

[0035] Figures 8a to 8d The changes on the substrate during the deposition of a functional material nanostructure array are shown.

[0036] Figure 9A and Figure 9B An SEM image of a cavity in a patterned photoresist layer on a substrate is shown according to an embodiment of this application.

[0037] Figure 10 Optical and SEM images of a CNT-modified DNA nanostructure template deposited on a substrate after the photoresist layer has been stripped, according to one embodiment of this application, are shown.

[0038] Figure 11 A graph showing the relationship between the angular distribution of aligned DNA nanostructures diffused into a PMMA cavity according to an embodiment of this application and the length of the DNA nanostructures is illustrated.

[0039] Figure 12 A multichannel CNT FET with ssDNA at the channel interface is shown. (A) A schematic diagram of the post-immobilization cleaning method is shown. (B) An AFM image magnified along the x and z projection directions of the CNT array after template removal is shown. Scale bar is 25 nm. See also Figure 16 and Figure 17 (C) A schematic diagram showing the design of introducing ssDNA at the channel interface and during FET fabrication. (D) A schematic diagram showing the I-channel DNA-containing CNT FET before thermal annealing (line I) and after thermal annealing (line II). ds -V gs The curve (plotted logarithmically at Vds at -0.5V). See also: Figure 20 .

[0040] Figure 13 The following diagram illustrates the fabrication of a top-gate high-performance CNT FET. (A) A schematic diagram of the design for fabricating a top-gate DNA-free FET. (B) Enlarged SEM images along the x and z projection directions show the multi-channel CNT FET used for fabrication. Dashed boxes indicate assembled CNT arrays. Scale bar is 100 nm. See also Figure 24 (C and D) show the Ii of single-channel (C) and multi-channel (D) CNT FETs. ds -V gs The curve (solid line, left axis, plotted logarithmically) and g m -V gs The curves (dashed lines, right axis, drawn linearly). Lines I, II, and III in C and D represent V values ​​of -0.8V, -0.5V, and -0.1V, respectively. ds See also Figure 22 and Figure 25 (E) shows benchmarking of the current multichannel CNT FET in D using other reports on high-performance CNT FETs. Previously published device performance (references 3, 5, 16-18, 23-27) is at -0.5V V. ds Obtained with channel lengths ranging from 100 nm to 500 nm. See also Figure 30 and Figure 31 .

[0041] Figure 14 The images show reduced-size (A) and magnified (B) TEM images of DNA-encapsulated CNTs. The scale bar in A is 200 nm. The scale bar in B is 100 nm.

[0042] Figure 15The height profiles of CNTs are shown. AFM images (A) and corresponding height profiles (B) of three different CNTs are presented. The dashed lines in (A) represent the locations of the height profiles in (B). The scale bar is 100 nm. As shown in the height map, the CNT diameter distribution ranges from less than 1 nm to approximately 1.5 nm.

[0043] Figure 16 SEM images of the immobilized CNT array after DNA removal are shown. In dashed box I, the two ends of the CNTs are fixed by two metal strips and used for FET construction. In the region of dashed box II, the unfixed CNT ends may have been disturbed during DNA removal and were therefore not used for FET construction. Scale bar is 500 nm.

[0044] Figure 17 AFM images of the CNT array fixed after DNA removal are shown. (A) A 3D scaled-down view of the CNT array fixed by two metal strips. (B) A magnified view of the CNTs fixed by metal rods. Scale bar: 25 nm. (C) More magnified AFM images of the CNT array fixed after DNA removal. Scale bar: 50 nm.

[0045] Figure 18 Schematic diagrams showing different compositions at the channel interface are shown. (A) shows the assembly and (B) shows the assembly after removal of the DNA template and metal ions.

[0046] Figure 19 A magnified SEM image of the constructed multichannel DNA-containing CNT FET is shown. Scale bar is 200 nm.

[0047] Figure 20 I is shown for multichannel DNA-containing CNT FETs. ds -V gs Curves. CNT FETs before (A) and after (B) thermal annealing. Different curves represent different CNT FETs. (C) A DNA-containing CNT FET in (A) under repeated measurements from 2V to -3V. Different curves represent different measurement values. V in (A), (B), and (C) ds All are set to -0.5V. I ds The spacing between CNTs is normalized.

[0048] Figure 21 The design of the constructed single-channel DNA-free CNT FET is shown. (A) Side view and (B) Top view of the FET design.

[0049] Figure 22 I is shown for all operable single-channel DNA-free CNT FETs. ds -V gsCurves. Different curves represent different CNT FETs. V ds Set to -0.5V.

[0050] Figure 23 The design of the constructed multichannel DNA-free CNT FET is shown. (A) Side view and (B) Top view of the FET design.

[0051] Figure 24 A scaled-down SEM image of the constructed multichannel DNA-free CNT FET is shown. Scale bar is 200 nm.

[0052] Figure 25 The I-type of all operable multichannel DNA-free CNT FETs is shown. ds -V gs Curves. Different curves represent different CNT FETs. (The last part, "I," appears to be a typo and can be left as is.) ds Relative to the normalization of the spacing between CNTs. V ds Set to -0.5V.

[0053] Figure 26 This demonstrates an I-type multichannel DNA-free CNT FET with the highest on-state current density at a channel length of 200 nm. ds -V ds Curves. Different curves represent different V values. gs V gs The range is -1.8V to 0.2V, with a step size of 0.2V. (The last part, "I," appears to be a typo and can be omitted.) ds The spacing between CNTs is normalized.

[0054] Figure 27 The transport performance of a multichannel DNA-free CNT FET with a channel length of 100 nm is shown. (A) shows the transport performance at -0.5 V. ds I ds -V gs The curve (left axis, plotted logarithmically) and g m -V gs Curve (right axis, plotted linearly). I ds and g m All are normalized relative to the spacing between CNTs. (B) shows I ds -V ds Curves. Different curves represent different V values. gs V gs The range is -1.4V to 0.6V, with a step size of 0.2V.

[0055] Figure 28 I is shown as a multichannel DNA-free CNT FET containing metal CNT impurities. ds -Vds Curve. V ds Set it to -0.5V. Set I... ds The spacing between CNTs is normalized.

[0056] Figure 29 The performance comparison of the constructed multichannel CNT FETs with different interface compositions is shown. From (A) to (E), the transconductance, subthreshold swing, threshold voltage, on-state conductance, and IT of the different FET samples are compared. on / I off Squares represent multi-channel DNA-containing CNT FETs before annealing. Circles represent heat-annealed multi-channel DNA-containing CNT FETs. Triangles represent multi-channel DNA-free CNT FETs. The sample number is the test number assigned to each FET. From (F) to (J), the transconductance, subthreshold swing, threshold voltage, on-state conductance, and It of different channel configurations are statistically analyzed. on / I off Bar I represents a multi-channel CNT FET containing DNA before annealing. Bar II represents a heat-annealed CNT FET containing multi-channel DNA. Bar III represents a multi-channel CNT FET without DNA. (At -0.5V...) ds Collect all performance data. For multichannel DNA-containing CNT FETs before and after annealing, at -3.0V V... gs Obtain its performance data. For multichannel DNA-free CNT FETs, at -1.5V V... gs Obtain performance data.

[0057] Figure 30 Benchmarks for CNT FETs with different CNT spacings are shown. Our multichannel CNT FETs have other reported benchmarks (even for CNT spacings) involving: (A), subthreshold swing, (B), transconductance (g) m () and (C), conductance in the on-state (G) on ). At -0.5V V ds The device performance previously published (References 3, 5, 15, 23) was obtained. Specifically, from Figure 4D in (Reference 23), Figures 4A and 4C in (Reference 3), and Figures 4D in (Reference 5) Figure 2 A and Figure 2 B and (Reference 15) Figure 3The transmission performance is obtained from Figures C and 4B. Channel lengths range from 100 nm to 500 nm. In each panel, the transmission performance (i.e., subthreshold swing, on-state conductance, and transconductance) is plotted against structural parameters (CNT spacing). High transmission performance requires simultaneously exhibiting a small subthreshold swing, high transconductance, and high on-state conductance. Compared to other FETs with different CNT spacings, our multi-channel CNT FET exhibits the smallest subthreshold swing, the highest transconductance, and the second highest on-state conductance.

[0058] Figure 31 Benchmarking of CNT FETs with different CNT densities is shown. Our multichannel CNT FET was benchmarked against other reports on high-density CNT arrays (with non-uniform CNT spacing), involving: (A), subthreshold swing, (B), transconductance (g m () and (C), conductance in the on-state (G) on The previously published device performance (references 16-18, 24-29) at -0.5V V ds The following can be obtained: Specifically, it can be obtained from Figure 4D in (Reference 17), Figure 1D in (Reference 18), Figure 1F in (Reference 16), and Figure 1F in (Reference 25). Figure 2 C. Figure 4A in (Reference 29), Figure 4A in (Reference 28) Figure 11 Figure 4B in (Reference 26) and Figure 4B in (Reference 27) Figure 2 B and Figure 2 These transmission performance characteristics were obtained from Figure 4C in (Reference 24). Channel lengths ranged from 100 nm to 500 nm. Transmission performance (i.e., subthreshold swing, on-state conductance, and transconductance) was plotted relative to structural parameters (CNT density) in each panel. High transmission performance requires a small subthreshold swing, high transconductance, and high on-state conductance simultaneously. Compared to other FETs with different CNT densities, our multi-channel CNT FET exhibited the second smallest subthreshold swing, the highest transconductance, and the third highest on-state conductance. Notably, the FET with the smallest subthreshold swing (Reference 27) exhibited an on-current density of less than 5 μA / µm, which does not meet the transmission requirements of a high-performance CNT FET.

[0059] Figure 32 A reduced-size TEM image of an assembled fixed-width CNT array with a CNT spacing of 16 nm is shown. The fixed-width DNA template exhibits a specified width at 34 nm. Arrows indicate assembled CNTs on the DNA template. The scale bar is 100 nm.

[0060] Figure 33SEM images of CNT-modified DNA templates aligned across 120 cavities are shown. Rectangles in the reduced-size SEM images indicate magnified locations. Arrows in the magnified SEM images indicate aligned DNA templates. Scale bar is 2 µm.

[0061] Figure 34 SEM images of a DNA template arranged within a rectangular PMMA cavity are shown. The PMMA cavity width is designed to be 2 μm with an aspect ratio of 1. The scale bar is 4 μm.

[0062] Figure 35 Different methods for fabricating CNT arrays with designer array widths, inter-array spacing, and CNT counts exceeding centimeter levels are illustrated. (A) shows processing a continuous CNT film (with random orientation) using a post-assembly etching step to generate designer array width / inter-array spacing / CNT count. (B) shows arranging a fixed-width CNT array (assembled using 3D DNA nanogrooves) within a pre-formed PMMA cavity, followed by PMMA stripping and DNA removal, which directly generates the geometry of the designer array without post-assembly etching. Detailed Implementation

[0063] In the following detailed description, reference is made to the accompanying drawings, which form a part of this specification. In the drawings, unless the context clearly indicates otherwise, similar reference numerals generally refer to similar parts. The illustrative embodiments described in the detailed description, drawings, and claims are not intended to be limiting. Other embodiments may be employed, and other changes may be made, without departing from the spirit or scope of the subject matter of this application. It should be understood that various configurations, alternatives, combinations, and designs of the various forms of this application, generally described herein and shown in the accompanying drawings, are intended to form part of this application.

[0064] Nanostructure templates (e.g., nucleic acid nanostructures) can be assembled into larger assemblies, and have been used to create complex shapes and extended crystals that may have periodic structures (including ribbon-like, tubular, circular, cubic, and other types of two-dimensional (2D) crystals, as well as various three-dimensional (3D) crystals or shapes). Such nanostructure templates, when modified or altered with functional nanostructures such as carbon nanotubes (CNTs), can form arrays of functional nanostructures, in which the functional nanostructures are precisely arranged. These functional nanostructures can be organized into nanodevices, circuits, etc. For example, U.S. Patent No. 7,842,793 provides a method for in vitro generation of nucleic acid nanostructures of arbitrary 2D or 3D shapes by folding a single-stranded DNA polynucleotide scaffold with multiple single-stranded oligonucleotide auxiliary strands according to a pre-specified folding path. Furthermore, “Yonggang Ke et al., DNA brick crystals with prescribed depths, Nature Chemistry, Vol. 6, 994-1002, November 2014” provides a general framework for constructing DNA brick crystals with prescribed depths and nanoscale 3D features (e.g., continuous or discontinuous cavities and channels that can be arranged to form functional nanostructures). U.S. Patent No. 10,099,920 provides a method for aligning nanoparticles and / or nanowires into nucleic acid nanostructures / crystals. The contents of these patent and non-patent references are incorporated herein by reference in their entirety. Since nanostructure templates are typically developed and formed in solution, it is desirable to arrange and fix individual nanostructure templates onto a substrate on which various subsequent nanofabrication processes can be applied to further scale up the size of the assembled nanostructure templates to the millimeter or centimeter level, or larger. "Ryan J. Kershner, Placement and Orientation of Individual DNA Shapes on Lithographically Patterned Surfaces, Nature Nanotechnology, Vol. 4, September 2009," describes a method for placing DNA shapes on a substrate by pre-forming a pattern with sites on the substrate surface. Figure 1a shows the dynamic binding process of DNA origami on the patterned surface, and Figures 1b and 1c show the alignment of the DNA origami on the substrate after the binding process. It can be seen, especially from Figures 1b and 1c, that due to the sites being wider than a single DNA origami, there are many random positions and orientations of DNA origami on the patterned surface, which is not desirable for industrial manufacturing, as these inaccurate arrangements and orientations can lead to a large number of defects.For example, the incorrect arrangement of DNA origami is particularly noticeable when the length of the DNA origami to be attached is uneven.

[0065] The inventors of this application have discovered that conventional methods for arranging nanostructure templates are unsatisfactory because they involve a trade-off between the yield of nanostructure template arrangement and the controllability of the orientation of the nanostructure template arranged on the substrate. Specifically, higher yields may require larger cavities to accommodate the nanostructure templates; however, this may loosen the cavity's constraints on the nanostructure templates, thus reducing the orientation accuracy of the nanostructure templates.

[0066] To address the aforementioned problems, embodiments of this application provide specific patterns in alignment layers that can efficiently and accurately arrange various pre-formed nanostructure templates on a substrate. In some embodiments, these patterned alignment layers have one or more cavities, each cavity comprising a main region for accommodating the nanostructure and at least one extension region extending from and in fluid communication with the main region for accommodating the nanostructure. The main region allows the nanostructure templates to diffuse into and be contained within the respective cavities on the substrate while substantially maintaining the pre-formed shape of the nanostructure templates. However, the extension regions cannot accommodate the entire structure of the pre-formed nanostructure templates while maintaining their pre-formed shape. For example, the extension regions can be configured to have a size smaller than that of the nanostructure template, such that the nanostructure templates cannot be contained within the extension regions. Furthermore, the extension regions can be shaped to have a shape different from and mismatched with the shape of the nanostructure template. As a result, the nanostructure templates cannot diffuse into and be contained within the extension regions. In other words, due to size limitations or shape mismatch, the extension regions cannot accommodate the entire structure of the nanostructure template. However, the extension regions can significantly increase the volume of the cavities, and thus increase the likelihood of containing the nanostructure templates during diffusion into the cavities. Furthermore, in some embodiments, the orientation of the nanostructure template can be precisely controlled by the shape and / or size of the host region accommodating the nanostructure. This significantly improves both yield and orientation controllability.

[0067] Figure 2 A method 200 for depositing nanostructures on a substrate according to one embodiment of this application is shown.

[0068] In some embodiments, the nanostructure template comprises one or more substances selected from the group consisting of: carbon nanotubes (CNTs), polymer-encapsulated CNTs, CNT films, semiconductor nanoparticles, semiconductor nanowires, semiconductor nanobulbs, metal nanoparticles, metal nanowires, metal nanobulbs, polymeric nanoparticles, polymeric nanowires, polymeric nanobulbs, ceramic nanoparticles, ceramic nanowires, ceramic nanobulbs, metal oxide nanoparticles, metal oxide nanowires, metal oxide nanobulbs, fluoride nanoparticles, fluoride nanowires, and fluoride nanobulbs. In some embodiments, the nucleic acid nanostructure template comprises one or more deoxyribonucleic acid (DNA) nanostructures, one or more ribonucleic acid (RNA) nanostructures, one or more locked nucleic acid (LNA) nanostructures, one or more peptide nucleic acid (PNA) nanostructures, or any combination of these nanostructures.

[0069] In some implementations, each nanostructure template can be modified with a functional nanostructure. The functional nanostructure comprises one or more substances selected from the group consisting of: carbon nanotubes (CNTs), polymer-encapsulated CNTs, CNT films, semiconductor nanoparticles, semiconductor nanowires, semiconductor nanoblocks, metal nanoparticles, metal nanowires, metal nanoblocks, polymeric nanoparticles, polymeric nanowires, polymeric nanoblocks, ceramic nanoparticles, ceramic nanowires, ceramic nanoblocks, metal oxide nanoparticles, metal oxide nanowires, metal oxide nanoblocks, fluoride nanoparticles, fluoride nanowires, and fluoride nanoblocks. The functional nanostructure can be considered as part of the nanostructure template.

[0070] like Figure 2As shown, method 200 begins at step 202, wherein a patterned alignment layer is formed on the surface of a substrate. In some embodiments, the substrate may include, but is not limited to, silicon, silicon dioxide (also known as silica), alumina, sapphire, germanium, gallium arsenide (GaAs), an alloy of silicon and germanium, or indium phosphide (InP). In some embodiments, the substrate may include silicon nitride, carbon, and / or polymers. In some embodiments, the substrate may be inorganic or organic. In some embodiments, the substrate may include graphene and / or graphite. In some embodiments, the substrate is a composite of any two or more materials (e.g., including mixtures) (e.g., a composite of inorganic and organic materials, or a composite of two or more different inorganic or organic materials). For example, the substrate may include a mixture of inorganic and organic materials, a mixture of two or more different inorganic materials, or a mixture of two or more different organic materials. In some embodiments, the substrate includes a semiconductor material or a mixture of semiconductor materials. Semiconductor materials include, but are not limited to, group IV semiconductors, group IV compound semiconductors, group VI semiconductors, group III-V semiconductors, group II-VI semiconductors, group I-VII semiconductors, group IV-VI semiconductors, group V-VI semiconductors, group II-V semiconductors, oxides, layered semiconductors, magnetic semiconductors, organic semiconductors, charge transfer complexes, and combinations thereof. Those skilled in the art will understand that the substrate can have a flat surface, a curved surface, or a surface with all the 3D features on which a patterned alignment layer can be formed. For example, some cavities of the patterned alignment layer may be higher than other cavities of the patterned alignment layer. Furthermore, if the substrate is placed horizontally, some substrates may have 3D features having vertically oriented surfaces, horizontally oriented surfaces (e.g., sidewalls), and / or inclined surfaces. Therefore, cavities can be formed on one or all of the vertically oriented surfaces, horizontally oriented surfaces, or inclined surfaces.

[0071] The patterned alignment layer has one or more cavities to guide the deposition of a nanostructure template on a substrate. The nanostructure template may have a generally predetermined shape and size, which typically does not change during deposition on the substrate. In some embodiments, the depth of each of the one or more cavities is two or more times, preferably five or more times, the thickness of the nanostructure template. Each cavity has a main region for accommodating at least one nanostructure template, and a plurality of extension regions extending from and in fluid communication with the main region. The extension regions have dimensions and shapes unsuitable for accommodating the entire structure of the nanostructure template, and therefore can only accommodate the nanostructure template within the main region. In some embodiments, a non-patterned alignment layer, such as a photoresist layer (e.g., polymethyl methacrylate (PMMA)), can be formed on the surface of the substrate. For example, a spin coating or spray coating process can be used to form the photoresist layer. The alignment layer can then be patterned using photolithography to form cavities with desired shapes and / or sizes at various locations on the substrate. In some embodiments, electron beam lithography can be used to pattern the alignment layer because finer patterns with resolutions below 10 nm can be formed. It is understood that other types of patterning processes can be used, such as photolithography, imprint lithography, and oriented self-assembly of block copolymers. In some embodiments, other types of materials, including polymers and inorganic materials, can be used as alignment layers. For example, polydimethylsiloxane (PDMS) or silicon oxide can be used as alignment layers.

[0072] Figure 3 An exemplary graphical alignment layer 300 on a substrate 302 is shown according to one embodiment of this application. (See attached image.) Figure 3As shown, the substrate 302 has a surface on which a patterned alignment layer 300 with six cavities 304 is formed. The number of cavities in the patterned alignment layer is for illustrative purposes only and is not intended to limit the scope of this application. Specifically, the upper left cavity 304 has a rectangular main region 306 and several extension regions 308 extending laterally from the periphery of the main region 306. Those skilled in the art will understand that the size and shape of the main region 306 can be determined according to the shape of the nanostructure template to be accommodated therein. For example, depending on what is to be accommodated, the main region 306 can have a circular shape, a triangular shape, a trapezoidal shape, a rectangular shape, etc. In some embodiments, the size and shape of the main region 306 can be the same as or larger than the size and shape of the nanostructure template, as long as the nanostructure template can be accommodated during the subsequent processes detailed below. In some embodiments, the extension regions 308 can have a smaller size than the main region 306 and / or the depth of the extension regions 308 can be equal to, greater than or less than the depth of the main region 306. In some embodiments, the extension regions 308 are formed together with the main region 306 during the patterning process, and therefore they can have substantially equal depths. The extension regions 308 increase the volume of the cavity 304. In some embodiments, the volume of the plurality of extension regions 308 in each cavity is 1%, 2%, 3%, 4%, 5% or more, preferably 10% or more, of the volume of the main region 306. In some embodiments, the area or volume of each extension region 308 can be from 1% to 30% of the area or volume of the main region 306.

[0073] exist Figure 3 In the illustrated embodiment, the extension region 308 has the same rectangular shape. In some other embodiments, the extension region may have other shapes, such as rhombus, triangle, spiral, circle, semicircle, trapezoid, or any other regular or irregular shape suitable for increasing the volume of the cavity but not for accommodating the nanostructure template. In some embodiments, the width or length of the extension region may be smaller than the width of the nanostructure template to be accommodated in the main region, such that the nanostructure template cannot be accommodated in the extension region. Figure 4a and Figure 4b An exemplary cavity with an extended region is shown, which may have other shapes or arrangements. Figure 4a As shown, eight cavities 400 are formed on a substrate 402. Each cavity 400 has a rectangular extension region 408 arranged in a comb-fingered manner. Furthermore, the extension regions 408 of every two adjacent cavities 400 can be inserted into each other at their adjacent sides but do not intersect. This reduces the spacing between two adjacent cavities 400, increasing the density of cavities 400 on the substrate 402. In some embodiments, the extension regions may have different shapes from each other. Figure 4b As shown, three cavities 450 are formed on a substrate 452. All cavities 450 have six sets of extension regions, and each set of extension regions includes a first folded extension region 458a and a second folded extension region 458b, the second folded extension region 458b being folded inside the first folded extension region 458a. The folded shape of the extension regions 458a and 458b can reduce the spacing between adjacent cavities 450, and thus increase the density of cavities 450 on the substrate 452. Those skilled in the art will understand that... Figure 4a The comb-like extension region shown and Figure 4b The folded extension region shown is exemplary and can be changed to other shapes, as long as the design and arrangement of the cavity can meet the minimum linewidth requirements provided by patterning processes such as electron beam lithography.

[0074] Refer again Figure 2 In step 204, a nanostructure template (e.g., a nucleic acid nanostructure) is diffused into one or more cavities of a patterned alignment layer. The nanostructure template may be contained in a pre-developed solution. For example, a multi-stage isothermal reaction can produce a solution containing the specified nucleic acid nanostructure. In one example, a mixture of unpurified DNA blocks, buffer material (e.g., tris(tris(hydroxymethyl)aminomethane)), nuclease inhibitors (e.g., EDTA (ethylenediaminetetraacetic acid)), and metal salts (e.g., MgCl2) for stabilizing the DNA blocks can be cultured according to a predetermined heating profile to develop nucleic acid block crystals into a nucleic acid nanostructure. In some embodiments, the nucleic acid nanostructure can be further modified or altered with functional nanostructures (e.g., CNTs), thus allowing for a suitable encapsulation process in the reaction buffer prior to step 204 to encapsulate the functional material nanostructure onto the nucleic acid nanostructure. For example, a solution containing the nucleic acid nanostructure can be mixed with a solution of NaCl or MgCl2 containing CNTs and subsequently cultured. More detailed examples of forming modified functional nanostructures of nucleic acid nanostructures can be found in U.S. Patent No. 10,099,920, the contents of which are incorporated herein by reference in their entirety. However, those skilled in the art will understand that any other type of nanostructure template, whether existing or to be developed in the future, with or without modification of functional nanostructures, can be formed using any desired method, and that such nanostructure templates can be deposited in cavities on a substrate, substantially retaining their pre-formed shape.

[0075] In some implementations, functional nanostructures, such as proteins, can be attached to nucleic acid nanostructures, thereby allowing for biological experiments on the complex assembly of proteins and the generation of molecular electronic or plasmonic circuits. Furthermore, metals (e.g., Ag, Au, Cu, or Pt), alloys (e.g., alloys comprising one or more of Ag, Au, Cu, Pt, or Si), semiconductors (e.g., Si, GaAs, InP, MoS2, TiO2), conductive metal oxides (e.g., In2O3, SnO2, Na2Pt3O4), and superconductors (e.g., Yba2Cu3O7, Ti2Ba2Ca2Cu3O7) can be used. 10 ) and / or magnetic nanoparticles or nanowires, fluorescent semiconductor quantum dots, or any other desired nanostructures can be attached as functional nanostructures to nucleic acid nanostructures. Depending on the post-processing of the substrate, these functional nanostructures can be used to form a variety of nanodevices, circuits, optical devices, etc.

[0076] In some embodiments, diffusion step 204 may further include impregnating the solution onto the patterned alignment layer and incubating the substrate. The impregnation and incubation steps may be performed simultaneously or sequentially. Incubating the substrate facilitates the diffusion of the nanostructure template into the cavities within the patterned alignment layer. For example, substrate incubation can be performed by dehydration or evaporation. In some embodiments, the substrate may be incubated in a sealed chamber for defined time intervals, such as 5 minutes, 10 minutes, 30 minutes, or longer, to allow sufficient time for the nanostructure template to firmly attach within the cavity. In some other embodiments, the substrate may be placed in an open environment or space to allow at least partial evaporation of the solvent. Diffusion step 204 may be implemented in other ways. For example, electrophoresis may be used to deposit the nanostructure template onto the substrate.

[0077] refer to Figure 3 The lower row includes three cavities 304, each filled with a nanostructure template 310. The nanostructure template 310 has substantially the same size and shape as the cavity 304. Thus, the nanostructure template 310 is constrained and oriented onto the substrate 302 as needed. In some embodiments, the nanostructure template 310 may include a previously attached array of functional nanostructures. For example, two parallel CNTs 312 are attached to the nanostructure template 310 in the lower left cavity 304. In this way, the two CNTs 312 can be precisely arranged and oriented into a functional nanostructure array, which further forms a larger nanostructure array with four other CNTs 312 in the lower middle and lower right cavities 304.

[0078] In some other embodiments, the size and shape of the main body region 306 can be larger than the size and shape of the nanostructure template to be arranged therein. For example, the length of the main body region 306 of the upper middle cavity is greater than the length of the nanostructure template 314, so the nanostructure template 314 can only move along the length of the main body region. Similarly, the width of the main body region 306 of the upper right cavity 304 is greater than the width of the nanostructure template 316, so the nanostructure template 316 can only move along the width of the main body region. It can be seen that although these main bodies can have larger dimensions because at least one translational degree of freedom of the nanostructure templates 314 and 316 along the surface of the substrate 302 is restricted, these main bodies can still orient the nanostructure templates 314 and 316. Thus, patterned alignment layers with such cavities show great promise in depositing nanostructure templates with non-uniform lengths, which is impossible with existing methods because these existing methods require precise matching of size and surface energy between the nucleic acid nanostructures and the patterned alignment layers.

[0079] although Figure 3 The width or length of the main body region of the cavity shown is approximately equal to the width or length of the nanostructure template. In some embodiments, the size and shape of the main body region may be larger than the nanostructure template. For example, the width and length of the main body region (e.g., rectangular) are larger than the width and length of the nanostructure template, respectively. Although such a size and shape of cavity may not completely orient the nanostructure template contained therein, the increased volume of the cavity increases the likelihood of the nanostructure template diffusing into the cavity.

[0080] The extension region 308 cannot accommodate at least one nanostructure template. In other words, the extension region 308, whether alone or collectively (even when bridged by the main region), cannot accommodate the entire structure of the nanostructure template. Thus, the nanostructure template can only be accommodated within the main region and is further restricted according to the shape of the main region. However, the extension region 308, which is in fluid communication with each of the main regions, can increase the overall volume of the cavity. In some embodiments, a cavity may have an extension region that can increase the volume of the cavity by at least 5%, or preferably by 10%, 20%, 30%, 50%, 100%, 150%, 200%, or more, compared to a cavity with only the main region. In some other embodiments, one or more other cavities may also have extension regions that similarly increase their respective volumes. As previously stated, the extension region 308 can significantly increase the volume of the cavity and thus increase the likelihood of accommodating the nanostructure template during diffusion into the cavity without reducing the orientation of the nanostructure template body contained therein.

[0081] Figure 5 Another exemplary graphical alignment layer 500 on a substrate 502 is shown according to one embodiment of this application. Similar to Figure 3 The cavities 304 shown in the patterned alignment layer 500 have multiple extended regions to increase their respective volumes, thereby improving the yield of nanostructure templates contained therein.

[0082] like Figure 5 As shown, four cavities are formed in the patterned alignment layer 500, and each cavity is filled with two nanostructure templates within its respective main body region. Specifically, the upper left cavity 504a is filled with two nanostructure templates 510a and 510b, whose lengths are shorter than the length of the main body region 506a of cavity 504a. However, the width of each of the two nanostructure templates 510a and 510b is essentially half the width of the main body region 506a, so their orientation is generally fixed and cannot be changed when they are housed parallel to each other within the main body region 506a. Similarly, the upper right cavity 504b is also filled with two nanostructure templates 510c and 510d, which are shorter than the main body region 506b. Unlike nanostructure templates 510a and 510b, nanostructure templates 510c and 510d are not aligned with each other in the length direction. However, if the functional nanostructures modifying these nanostructure templates 510c and 510d are arranged longitudinally, such as... Figure 3 As shown in CNT 312, this misalignment may not significantly affect the further fabrication of the nanodevice or circuits based thereon. In some preferred embodiments, the main body region of the cavity in the patterned alignment layer can be precisely designed to accommodate a desired number and size of nanostructure templates. For example, the length of the lower left cavity 504c can be substantially equal to the length of the nanostructure templates 510e and 510f, and the width can be substantially twice the width of the nanostructure templates 510e and 510f. In this way, the assembled nanostructures 510e and 510f are confined within the main body region 504c without any degrees of freedom along the surface of the substrate 502.

[0083] Figure 6 Another exemplary graphical alignment layer 600 on a substrate 602 is shown according to one embodiment of this application. Figure 6 As shown, the alignment layer 600 has a cavity 604 for accommodating a nanostructure template 610, which is shaped like a triangular ring. The main body region 606 of the cavity 604 is similarly triangular in shape, conforming to the contour of the nanostructure template 610. Furthermore, multiple parallelogram or triangular extension regions 608 extend from three sides of the triangular main body region 606 to respectively increase the volume of the cavity 604.

[0084] Figure 7 Another exemplary graphical alignment layer 700 on a substrate 702 is shown according to an embodiment of this application. (See attached diagram.) Figure 7 As shown, the alignment layer 700 has two cavities 704, each cavity for accommodating four triangular nanostructure templates 710 assembled together. The main body region 706 of the cavity 704 has a triangular shape that conforms to the contour of the four assembled nanostructure templates 710. Furthermore, multiple parallelogram or rectangular extension regions 708 extend from three sides of the triangular main body region 706 to increase the volume of the cavity 704. Because the extension regions 708 can help the nanostructure templates 710 accumulate and diffuse into the cavity 704, it is easier to assemble the triangular nanostructure templates 710 into larger triangular shapes as needed.

[0085] Refer again Figure 2 In some embodiments, prior to step 204, method 200 may include differentiating the substrate to improve the adhesion of the substrate surface to the nanostructure template. For example, dry oxidation etching of a substrate with a patterned alignment layer may be used to treat the surface of the substrate exposed through the cavity. It is understood that various differentiation processes may be used in the differentiation step. For example, in some embodiments, physical differentiation may also be used, such as glow discharge of the substrate, i.e., bombarding the substrate with oxygen plasma.

[0086] Following step 204, in which the nanostructure template is diffused into the cavity of the patterned alignment layer, the patterned alignment layer can be removed from the substrate in step 206, leaving only the nanostructure template attached to the substrate. In some embodiments, a stripping process can be used to remove the patterned alignment layer. Those skilled in the art will understand that in step 204, some nanostructure template may be attached to the top of the patterned alignment layer. Thus, these portions of the nanostructure template can also be removed along with the patterned alignment layer.

[0087] Figure 8a and Figure 8b Exemplary bases 802 with and without graphical alignment layers are shown respectively. Figure 8a and Figure 8b As shown, after removing the patterned alignment layer 800, the remaining nanostructure template 810 forms a nanostructure array that maintains the original positional relationships defined by the removed patterned alignment layer 800. Furthermore, each nanostructure template 810 is modified with two functional nanostructures 812 (e.g., CNTs). In some embodiments, such as Figure 8dAs shown, the non-functional portions of the nanostructure template 810 (i.e., those other than the functional nanostructure) can be further removed to leave only the functional nanostructure 812 on the substrate 802. For example, when the nanostructure template 810 is a nucleic acid nanostructure, the nucleic acid molecules of the nucleic acid nanostructure can be removed by an oxidizing agent solution (e.g., hydrogen peroxide solution). Preferably, prior to the removal of the nucleic acid molecules, certain immobilization structures can be formed on the substrate 802 to immobilize the nucleic acid nanostructure 810 (especially the functional nanostructure 812) on the substrate 802. For example, as... Figure 8c As shown, metal strips, polycrystalline silicon strips, or other inorganic material strips 811 can be formed covering the functional nanostructure 812, while simultaneously anchoring to the surface of the substrate 802 that does not cover the nucleic acid nanostructure, such as... Figure 8c As shown. When the nucleic acid nanostructure 810 is removed, the immobilized structure can remain on the substrate. For more details on the immobilized structure, please refer to PCT application PCT / CN2020 / 082375, entitled "Method for forming nanostructures and field-effect transistor devices on a substrate", filed March 31, 2020, and PCT application PCT / CN2020 / 082778, entitled "Method for forming nanostructures and field-effect transistor devices on a substrate", filed April 1, 2020, the entire contents of which are incorporated herein by reference.

[0088] It is understood that in some embodiments, the nanostructure template attached to the substrate is not modified with a functional nanostructure, and the template portion of the nanostructure template itself (e.g., a nucleic acid nanostructure) is the nanostructure required to form the nanostructure array. Thus, it may not be necessary to remove the template portion of the aforementioned nanostructure template, and some other post-processing can be performed on the nanostructure template on the substrate. Similarly, when both the template portion and the functional nanostructure are required, it may not be necessary to remove the template portion. In some embodiments, either or both of the template portion or the functional nanostructure may include two or more substances, and removal of one or more substances from the template portion and the functional nanostructure can be performed. For example, the functional nanostructure may be a DNA-encapsulated CNT, which can be attached to the nucleic acid nanostructure template, and the DNA encapsulating the CNT can be removed along with the nucleic acid nanostructure template, thereby leaving only the CNT on the substrate.

[0089] Example 1

[0090] An array of sample nanostructures composed of nucleic acid nanostructures has been produced using the method according to the embodiments of this application. Specifically, a 0.35 cm 2A silicon substrate (covered with a 300nm thick SiO2 layer) was spin-coated with PMMA photoresist (Allresist GmbH model: AR-P 672.045). Electron beam lithography (Raith GmbH model: Voyager, with an exposure dose of 325uC / cm at a current of 0.9nA) was used. 2 The PMMA photoresist is patterned to form the desired cavity. The patterned PMMA photoresist layer is then developed in a developing solution (a 1:3 mixture of methyl isobutyl ketone (MIBK) and isopropanol (IPA)) to remove exposed portions of the PMMA photoresist. Afterward, the substrate is rinsed with IPA and dried with nitrogen. Figure 9A and Figure 9B SEM images of cavities within a patterned PMMA photoresist layer on a substrate are shown. Figure 9A and Figure 9B As shown, each cavity has an elongated main body region and multiple extension regions extending laterally from the main body region. The scale is [not specified]. Figure 9A The middle is 40μm, in Figure 9B The thickness is 2 μm. After patterning, the patterned PMMA photoresist layer in a write field in SEM imaging has a thickness of 5 × 10⁻⁶. 4 One PMMA cavity (surface density approximately 2 × 10⁻⁶) 7 Each cavity / cm 2 Each PMMA cavity is 2.5 μm long, 180 to 250 nm wide, and 150 nm deep. It can be seen that the extended region significantly increases the cavity volume.

[0091] Next, a solution (5 μL) of CNT-modified DNA nanostructures was impregnated onto a patterned PMMA photoresist, and the substrate was then placed in a sealed chamber for 2 hours. Each CNT-modified DNA nanostructure comprised an assembled array of fixed-width CNTs with a specified CNT spacing of 16 nm and two CNTs. During this process, the DNA nanostructures diffused into the cavities of the patterned PMMA photoresist layer. The substrate was then dried, followed by the removal of the PMMA in acetone, leaving only the DNA nanostructures that had previously diffused into the cavities on the substrate. Figure 10 Optical and SEM images of CNT-modified DNA nanostructures formed on the substrate after the PMMA photoresist layer was stripped are shown. Figure 10 As shown, the image in the upper left corner is 0.35cm in size. 2Optical images of the silicon substrate. The remaining three images are reduced and magnified (rectangles show selected areas used for magnified views) to show SEM images of the silicon substrate with DNA nanostructures. The scale bars are 10 μm in the lower left image, 1 μm in the middle image, and 500 nm in the right image. The arrows in the right image indicate the DNA nanostructures. It can be seen that the DNA nanostructures are precisely arranged and oriented as needed (relative to...). Figure 9A and Figure 9B (The cavity arrangement is shown).

[0092] In SEM-based counting, over 85% (approximately 600 cavities) of the initial PMMA cavities in the patterned PMMA photoresist layer were occupied by CNT-modified DNA nanostructures. The angular distribution was defined as the difference between the longitudinal axis of the DNA nanostructure and the longitudinal direction of the PMMA cavity on the substrate. For all remaining DNA nanostructures within the occupied 600 cavities, the measured angular distribution was within ±1° for 56% and within ±7° for 90%. The measured angular distribution incorporated the effects of fabrication defects in the PMMA cavities, variations during DNA placement, and interference during PMMA stripping. Notably, the angular distribution was significantly improved when deploying DNA template inorganic materials on a large scale compared to previous reports (see AMHung et al., Large-area spatially ordered arrays of gold nanoparticles directed by lithographically confined DNA origami, Nature Nanotech. 5, 121-126 (2010)).

[0093] The length of the DNA nanostructure and the aspect ratio of the PMMA cavity can both affect the angular distribution. Compared to the angular distribution of shorter DNA nanostructures (length < 500 nm, 1° ± 11°), longer DNA nanostructures (length > 1 μm) may exhibit a narrower angular distribution (0° ± 3.4°). Figure 11A graph showing the relationship between the angular distribution of aligned DNA nanostructures diffused into PMMA cavities and the length of the DNA nanostructures is illustrated. Furthermore, PMMA cavities with a higher aspect ratio (e.g., 10 or higher) can provide better orientation controllability compared to PMMA cavities with a lower aspect ratio (e.g., 1 to 2 or less). Therefore, longer DNA nanostructures and higher aspect ratios of the PMMA cavities may be beneficial for further improving the angular distribution. In some embodiments, the anisotropic nucleic acid nanostructures (e.g., elongated nucleic acid nanostructures) to be deposited on the substrate can have a length of 500 nm or longer, or preferably 1 μm, 2 μm, 5 μm or longer. Additionally, in some embodiments, the aspect ratio of the cavities within the patterned alignment layer can be 2 or higher, or preferably 5, 10, 20, 30 or higher.

[0094] Typically, using the methods according to embodiments of this application, anisotropic biological or non-biological nanostructure arrays can be aligned and attached to pre-formed cavities on a centimeter-scale substrate. In some preferred embodiments, effective spatial confinement and orientation of the nanostructure array can be achieved, applicable to both regular and irregular nanostructures.

[0095] Example 2

[0096] Example 2 is an example of constructing a high-performance transistor using the method according to the embodiments of this application.

[0097] In planned high-performance energy-saving field-effect transistors (FETs) (References 1, 2), uniformly spaced, small-pitch semiconductor channels (the spacing between two adjacent channels in a single FET) are typically required. Smaller channel spacing leads to higher integration density and on-state performance, but carries the risk of enhanced destructive short-range shielding and electrostatic interactions in low-dimensional semiconductors (e.g., carbon nanotubes (CNTs)) (Reference 3); while uniform alignment minimizes channel disorder affecting the switching between on / off states (Reference 4). Therefore, although high-density CNT films exhibit on-state performance comparable to silicon FETs (References 5, 6), reduced gate modulation and increased subthreshold swing are observed due to disorder in the array (References 3, 5).

[0098] Biomolecules such as DNA (References 7, 8) can be used to organize CNTs into prescribed arrays (References 9-11). Based on space-constrained integration (SHINE) of nanotube electronics, biofabrication has further reduced the uniformly spaced channel pitch, making it beyond the feasibility of photolithography (Reference 12). However, no biotemplated CNT FET (References 12-14) has exhibited performance comparable to CNT FETs constructed by photolithography (Reference 15) or thin-film methods (References 3, 5, 6, 16-18). Furthermore, the wide orientation distribution (Reference 19) during the surface arrangement of biotemplate materials prevents their large-scale alignment.

[0099] Here, we demonstrate that small regions of nanoscale biomolecular assemblies can be integrated into large arrays of solid-state high-performance electronic devices. We use a parallel semiconductor CNT array assembled via SHINE as a model system (Reference 12). At the FET channel interface, we observe a decrease in on-state performance caused by high concentrations of DNA / metal ions. Using a post-fixation cleaning method, we eliminate contamination without degrading CNT alignment. Based on uniform CNT spacing and a clean channel interface, we construct a solid-state multi-channel PMOS (p-channel metal-oxide-semiconductor) CNT FET that exhibits both high on-state performance and fast on / off switching. Using a photolithographically defined polymethyl methacrylate (PMMA) cavity to spatially define the position of the CNT-modified DNA template, we demonstrate a 0.35 cm... 2 An array aligned with a prescribed geometry on a substrate of area. Constructing high-performance, ultra-large-scale devices on bioelectronic interfaces enables a variety of post-silicon applications, such as multiplexed biomolecular sensors with nanometer-to-centimeter array scalability (Reference 20) and 3DFETs.

[0100] We assembled DNA-templated CNT arrays using DNA-based SHINE (Reference 12). We applied a post-fixation washing method ( Figure 12 A) To remove the DNA template. Starting with the CNT array of the DNA template deposited on the surface, the two ends of the CNT array of the DNA template are first fixed to the silicon wafer by deposited metal strips. Figure 12 The first step in A). High concentrations of metal salts (1 to 2 M) within the DNA template and DNA helix are gently removed by sequential washing with water and low-concentration H2O2. Figure 12 A and Figure 18 The second step in the process. During cleaning, the CNT spacing and the alignment quality of the assembled CNTs did not decrease ( Figure 12 B. Figure 16 and Figure 17 ).

[0101] To explore the effect of single-stranded DNA (ssDNA) at the channel interface, we first fabricated the source and drain electrodes on a cleaned CNT array. Figure 12 C, left). Next, ssDNA will be specifically introduced into the predetermined channel region ( Figure 12 In the first step of C, the channel length is 200 nm. Finally, the HfO2 gate dielectric and the Pd gate electrode are fabricated sequentially. Figure 12 The second and third steps in C and Figure 19 ).

[0102] Of the 19 FETs we constructed, 63% (12 out of 19) exhibited typical gate modulation (I0). on / I off More than 10 3 , Figure 20 ). The I of the other 7 devices on / I off <5, this is caused by the presence of metal CNTs within the array. At a source-drain bias of -0.5V (V ds Under these conditions, a typical multichannel DNA-containing CNT FET ( Figure 12 D) shows the threshold voltage (Vth) near -2V and the gate-source bias (Vth) at -3V. gs The on-state current density is 50 μA / μm (normalized relative to the CNT spacing), the subthreshold swing is 146 mV per decade (mV / decade), and the peak transconductance (g) is... m The conductance is 23 μS / μm, and the conductance in the on-state (G) is... on The value is 0.10 mS / μm. Statistics for all 12 operable FETs show a VF of -2 ± 0.10 V. th Distribution, conduction current density of 4-50 μA / μm and subthreshold swing of 164±44 mV per decibel ( Figure 20 A). Transmission performance remained stable during repeated measurements. Figure 20 C).

[0103] We annealed the aforementioned DNA-containing FETs at 400°C for 30 minutes under vacuum to thermally decompose the ssDNA (Reference 22), and then recharacterized their transport performance. Compared with the unannealed samples, the thermally annealed ( Figure 12 D、 Figure 19 and Figure 29 Slightly deviated from the average V th (Approximately 0.35V, V after annealing) thThe value was -1.65 ± 0.17 V, and the average subthreshold swing increased by approximately 70 mV per decade (the subthreshold swing after annealing was 230 ± 112 mV per decade). After annealing, including g m and G on Other conduction states and FET configurations remain largely unchanged.

[0104] To construct high-performance CNT FETs from biological templates, we deposited a composite gate mediator (Y₂O₃ and HfO₂) into the cleaned channel region instead of introducing ssDNA. Figure 13 A and B Figure 23 and Figure 24 Of all the FETs constructed, 54% (6 out of 11) exhibit gate modulation. Figure 25 The other five of the 11 FETs contain at least one metal CNT within the channel. Figure 28 Using the same fabrication process, we also built nine additional operational single-channel DNA-free CNTFETs for comparing transport performance. Figure 21 At the subthreshold swing hot electron limit (i.e., 60 mV per decade of range), Figure 13 C and Figure 22 Under these conditions, the single-channel CNT FET (channel length ~200nm) exhibits the highest on-state performance with an on-current of 10μA / CNT (Vds = -0.5V).

[0105] At -0.5V ds At this point, it has the highest conduction performance ( Figure 13 D and Figure 26 The multichannel DNA-free CNT FET (channel length ~200 nm, CNT spacing 24 nm) exhibited a Vt of -0.26 V. th The conduction current density is 154 μA / μm (at -1.5V). gs (Time), and the subthreshold swing is 100mV per decade. g m and G on The values ​​were 0.37 mS / μm and 0.31 mS / μm, respectively. g m -V gs The noise in the curve may originate from thermal noise, disorder, and scattering within the composite gate structure. In V ds At -0.8V, the on-state current further increases to ~250μA / μm, g m The value is 0.45 mS / μm, and the subthreshold swing is 110 mV per decade.

[0106] When the channel length is 100 nm, we achieved a conduction current density of 300 μA / μm (at -0.5 V). ds and -1.5V of V gs (time), and subthreshold swing of 160mV per decade ( Figure 27 Therefore, G on and g m The values ​​were all increased to 0.6 mS / μm. DNA-free CNT FETs exhibited Ig values ​​comparable to thin-film FETs of aligned chemical vapor deposition (CVD) grown CNT arrays (References 28, 29). ds Even with a 60% lower CNT density (~40 CNTs / μm compared to greater than 100 CNTs / μm in (References 28, 29)), effective removal of contaminants (e.g., DNA and metal ions) and shorter channel lengths contribute to improved I ds It is worth noting that previous studies directly attached CNTs to the source and drain electrodes (Reference 13), but due to the inability to completely remove contaminants from the electrode contact area, the on-state performance (g) was affected. m and G on It decreased by 10 times.

[0107] With similar channel length and V ds (i.e., -0.5V) We used CVD-grown or polymer-encapsulated CNTs to compare the current transport performance (i.e., g) of conventional thin-film FETs. m Benchmark tests were conducted on (and subthreshold swing) (References 3, 5, 16-18, 23-27) Figure 13 E, Figure 30 and Figure 31 High on-state performance (approximately 0.37 mS / μm g) can be simultaneously achieved in the same solid-state DNA template FET. m ) and fast on / off switching (approximately 100 mV per decibel of range subthreshold swing); while thin-film CNT FETs with similar subthreshold swing (approximately 100 mV per decibel of range) g m Reduced by more than 50% Figure 30 ).

[0108] Furthermore, the subthreshold swing difference between multi-channel (average 103 mV per decibel) and single-channel CNT FETs ( Figure 22 The average value (86 mV / decathlon) decreased to 17 mV per decibel. Theoretical simulations show that, under the same gate configuration, CNT diameter inhomogeneity (Reference 6) and alignment disorder (including intersecting CNTs) (Reference 5) increase the subthreshold swing (Reference 4). We use AFM images ( Figure 15 ) and TEM images ( Figure 14 A wide diameter distribution of DNA-encapsulated CNTs was observed in the study. Therefore, the aforementioned small subthreshold swing difference indicates effective gate modulation and uniformly spaced CNT alignment using SHINE (Reference 12), i.e., the absence of cross-branched / split CNTs in the channel region.

[0109] Statistical data for all operable multichannel DNA-free FETs showed a VF of -0.32 ± 0.27 V. th 25 to 154 μA / μm conduction current density (at -0.5 V) ds and -1.5V of V gs The current density exhibits a subthreshold swing of 103 ± 30 mV per decibel. Different numbers of narrow CNTs (i.e., diameter < 1 nm) within the FET result in a wide distribution of conduction current density. Since the Schottky barrier and band gap increase with decreasing CNT diameter, the observed conductivity of carbon nanotubes is typically lower than that of carbon nanotubes with diameters greater than 1.4 nm (References 30, 31).

[0110] When comparing the transmission performance differences between DNA-containing FETs and DNA-free FETs ( Figure 29 We observed V th A significant negative shift (-2V relative to -0.32V) occurs at positive V. gs higher I ds (Mostly 10 to 200 nA / μm relative to 0.1 to 10 nA / μm), and more than an order of magnitude smaller g m (4 to 50 μS / μm vs. 70 to 370 μS / μm). Therefore, high concentrations of ssDNA within a multichannel FET impair transport performance. Thermal annealing cannot completely eliminate this effect due to the presence of insoluble annealing products, such as metal phosphates (Reference 22).

[0111] When CNT-modified DNA templates are deposited on a flat silicon wafer, unrestricted surface rotation results in random orientation of the DNA templates. We address this issue by using 3D polymer-type cavities to define the surface orientation during large-area deployment. We first assembled fixed-width CNT arrays with a specified CNT spacing of 16 nm (2 CNTs per array). Figure 32 Next, in a typical 500μm x 500μm write field (at 0.35cm) on a PMMA-coated silicon substrate. 2 (With more than 20 write fields on the substrate), we fabricated densely aligned, serrated, wall-like PMMA cavities (cavity density approximately 2 × 10⁻⁶). 7 cavity / cm2 The minimum and maximum design widths along the z-direction are 180 nm and 250 nm, respectively.

[0112] After DNA deposition and PMMA stripping ( Figure 10 >85% of the initial cavities (counted to approximately 600 cavities) were occupied by the DNA template. Figure 10 , Figure 22 Based on the counts from each scanning electron microscope (SEM) scan, all remaining DNA templates within 600 cavity locations were counted. The measured angular distribution (defined as the difference between the longitudinal axis of the DNA template and the x-direction of the substrate) was 56% within ±1° and 90% within ±7°. Figure 10 This value includes the mitigateable effects of manufacturing defects in the PMMA cavity location, variations during DNA placement, and any interference from PMMA stripping. Notably, the angular distribution was still improved compared to previous large-scale placements of DNA template material (Reference 19). Because CNTs are embedded within the DNA grooves and shielded from the SEM detector by the DNA helix, CNTs are not visible under SEM.

[0113] The length of the DNA template and the aspect ratio of the PMMA cavity both affect the angular distribution. The angular distribution is different for shorter DNA templates (length <500nm). Figure 10 Compared to the mean (1°±11°), longer DNA templates (length>1μm) have a narrower angular distribution. Figure 10 The midpoint is 0°±3.4°. Additionally, it has a high aspect ratio (i.e., Figure 10 The PMMA cavity of 10) has a low aspect ratio (i.e., Figure 34 1) The PMMA cavity offers better orientation control. Therefore, a longer DNA template and a higher aspect ratio of the PMMA cavity are beneficial for further improving the angular distribution. Because the PMMA cavity is wider than the DNA template, we observed up to three DNA templates within several PMMA cavities, as well as offsets of the DNA template along the x and z directions. Notably, even with a saturated DNA solution, the DNA template does not completely cover the PMMA cavity.

[0114] Two-dimensional hydrophilic surface patterns with the same shape and size as DNA structures can guide the orientation of deposited DNA structures (Reference 32). However, it is difficult to design patterns that adapt to DNA templates with variable lengths. Instead, effective spatial constraints depend primarily on the length of the DNA template and the aspect ratio of the PMMA cavity, and are suitable for irregular template lengths. Therefore, anisotropic biotemplated CNT arrays can be arranged along the longitudinal direction of the cavity (…). Figure 35 ).

[0115] Reducing the CNT spacing to below 10 nm could be beneficial for further improving on-state performance. However, at a CNT spacing of 2 nm, enhanced electrostatic interactions may affect on / off switching. Therefore, the correlation between CNT spacing and CNT FET performance metrics needs to be verified. Combined with large-area fabrication via conventional photolithography and the directional assembly of bulk copolymers, biomolecular assembly can provide a high-resolution paradigm for large-area programmable electronics. Composite electronic biodevices can also integrate electrical stimulation and biological inputs / outputs, resulting in ultra-large-scale sensors or bioactuators.

[0116] Materials and Experimental Methods in Example 2

[0117] 1. Atomic Force Microscopy (AFM)

[0118] The prepared 7 μL CNT-modified DNA template solution was deposited on a 1 cm² plate. 2 The samples were then progressively cleaned on silicon wafers in 50%, 95%, and 99.5% ethanol. The samples were imaged onto a multi-mode SPM (Vecco) using a tapping mode.

[0119] 2. Scanning electron microscope (SEM)

[0120] The prepared 7 μL CNT-modified DNA template solution was deposited on a 1 cm² plate. 2 The silicon wafers were then progressively cleaned in 50%, 95%, and 99.5% ethanol. The dried silicon chips were then imaged on a HITACHI S-4800 system operating under a high vacuum of 5 kV.

[0121] 3. Transmission electron microscopy (TEM)

[0122] 0.6 μL (unpurified) of the prepared CNT-modified DNA template was diluted in 5 μL of water and adsorbed onto a glow-discharged carbon-coated TEM grid for 4 min. Residual solution was then wiped away, followed by negative staining (7 s) with 6 μL of 2% uranyl formate aqueous solution and rapid washing. Imaging was performed using a JEOL 2100 operating at 120 kV.

[0123] 4. Centimeter-level orientation arrangement

[0124] First, spin-coating a polymethyl methacrylate (PMMA) resist (Allresist AR-P 672.045) to a size of 0.35 cm was performed. 2 The silicon substrate was subjected to electron beam lithography (Raith Voyager) at an exposure dose of 325 uC / cm at a current of 0.9 nA.2 The patterned PMMA layer was developed in a 1:3 mixture of methyl isobutyl ketone (MIBK) and isopropanol (IPA), then washed with IPA and dried with nitrogen. A CNT-modified DNA template solution was impregnated onto the photolithographically defined pattern. The silicon substrate was then held in a sealed chamber for 2 hours. During this process, the DNA template diffused into the PMMA chamber. The silicon substrate was then dried, followed by PMMA stripping, leaving only the aligned DNA template on the flat silicon substrate. Finally, the sample was imaged using SEM.

[0125] 5. Remove DNA template

[0126] We used the following process to remove the assembled DNA template while maintaining CNT alignment: (1) alignment marks were fabricated on a silicon wafer using electron beam lithography; (2) the CNT-modified DNA template was deposited on the silicon wafer, and its position was recorded using low-magnification SEM; (3) metal strips were fabricated to fix the assembled CNT array onto the silicon wafer; and (4) the DNA template was removed by continuous washing with water and H2O2. We used CNTs (semiconductor purity >95%) from NIST, classified by length, ranging from 300 to 1000 nm.

[0127] Alignment mark:

[0128] A 230 nm thick PMMA layer was spin-coated onto a silicon wafer (with a 300 nm thick SiO2 layer on top), and the Raith Voyager system was used (at a current of 9 nA and a temperature of 780 μC / cm). 2 A fine alignment mark pattern was written at a specific dosage. The alignment mark pattern was developed in a 1:3 mixture of MIBK and IPA. Stacked titanium / gold films (5 nm thick titanium and 45 nm thick gold) were deposited using a DE400 electron beam evaporation system. The films were then peeled off in acetone at room temperature without sonication and washed with ethanol. The samples were dried with nitrogen.

[0129] CNT deposition and recording:

[0130] A 9 μL solution of assembled CNT-modified DNA template was immersed in an oxygen plasma-cleaned labeled silicon wafer and incubated at room temperature for 1 hour. Afterward, the remaining solution was purged with nitrogen. The silicon wafer was then washed sequentially with 75%, 95%, and 99% ethanol and air-dried. The silicon wafer was then imaged at low magnification (operated at 1 kV) under SEM. The positions of the CNT-modified DNA template were recorded relative to the alignment markers.

[0131] CNT fixation and DNA removal:

[0132] A 230 nm thick PMMA layer was spin-coated onto a CNT-deposited silicon wafer. This was done using the Raith Voyager system (at a current of 400 pA and a temperature of 750 μC / cm). 2 Metal strip patterns were written at a specific dosage. The metal strip patterns were developed in a 1:3 mixture of MIBK and IPA. A stacked film of 5 nm thick titanium and 60 nm thick gold was deposited using a DE400 electron beam evaporation system. The film was stripped in acetone at room temperature without sonication, followed by washing with ethanol. The sample was dried with nitrogen. DNA removal was then performed by sequential washing with water and H2O2 (5%).

[0133] 6. FET Construction

[0134] For FET fabrication, we use electron beam lithography to fabricate source / drain / gate electrodes onto the assembled CNT array and construct electrical contact pads.

[0135] Source electrode / drain electrode:

[0136] A 230 nm thick PMMA layer was spin-coated onto a clean CNT array, and then the Raith Voyager system was used (at a current of 400 pA and a current of 750 uC / cm). 2 Source and drain electrode patterns were written at a specific dosage. The source and drain electrode patterns were developed in a 1:3 mixture of MIBK and IPA. A stacked film of 0.5 nm titanium, 30 nm palladium, and 40 nm gold was deposited using a DE400 electron beam evaporation system. The film was stripped in acetone at room temperature without sonication, followed by washing with ethanol. The sample was dried with nitrogen.

[0137] Gate electrode:

[0138] Next, a 230 nm thick PMMA layer was spin-coated onto a silicon wafer, and then a channel pattern was written using the Raith Voyager system (at a current of 400 pA and a dose of 750 uC / cm²). First, a 1 nm thick yttrium metal film was deposited using a DE400 electron beam evaporation system. This was followed by stripping in acetone at 70 °C. The yttrium film was then oxidized in air at 250 °C.

[0139] Then, a 230nm thick PMMA layer was spin-coated onto a silicon wafer covered with Y2O3, and then the Raith Voyager system was used (at a current of 400pA and a temperature of 750uC / cm). 2A gate electrode pattern was written at a specific dosage. The gate electrode pattern was developed in a 1:3 mixture of MIBK and IPA. Next, an 8 nm thick HfO2 film was deposited by atomic layer deposition (Beneq) at 90 °C. Finally, a 15 nm thick palladium film was deposited using a DE400 electron beam evaporation system. The film was stripped in acetone at room temperature without sonication and then washed with ethanol. The sample was dried with nitrogen.

[0140] Contact pad:

[0141] To fabricate larger electrical contact pads connected to the electrodes, a 230 nm thick PMMA layer was first spin-coated onto the sample. The contact pad pattern was exposed using a Raith Voyager system (at a current of 9 nA and a dose of 750 uC / cm²). The contact pad pattern was developed in a 1:3 mixture of MIBK and IPA and then dried with nitrogen. A stacked film of 5 nm thick titanium and 70 nm thick gold was deposited using a DE400 electron beam evaporation system. The film was then peeled off in acetone at room temperature without sonication and washed with ethanol. The sample was then dried with nitrogen.

[0142] Electrical measurements of CNT FETs:

[0143] Electrical measurements of the constructed CNT FET were performed at room temperature in a probe station connected to a Keithley 4200SCS semiconductor device analyzer.

[0144] 7. Introduce ssDNA at the channel interface

[0145] After fabricating the source / drain electrodes, we applied the following process to introduce ssDNA at the channel interface and accordingly construct the gate dielectric: (1) spin-coating a 230 nm thick PMMA layer onto the wafer, and then using the Raith Voyager system (at a current of 400 pA and 750 uC / cm) 2 (1) At a dose of 10 μL, a gate electrode pattern was written. The gate electrode pattern was developed in a 1:3 mixture of MIBK and IPA; (2) 10 μL of L1 solution (1 μM) was immersed in the fixed CNT array and incubated at room temperature for 1.5 h; (3) The remaining solution was blown away with nitrogen and then washed sequentially with 75%, 95% and 99% ethanol; (4) A 9 nm thick HfO2 medium was grown in the developed pattern by atomic layer deposition (Savannah) at 90 °C. A 15 nm thick palladium film was deposited using a DE400 electron beam evaporation system. The sample was peeled off in acetone at room temperature without sonication and then washed with ethanol. The sample was dried with nitrogen.

[0146] Then, perform contact pad and electrical measurements using the same method described in the “FET Construction” section above.

[0147] Further optimization of FET performance in Example 2

[0148] To further improve FET performance, it is necessary to increase the on-state conductance while reducing the subthreshold swing.

[0149] Several strategies have been proposed in previous reports to improve on-state conductance. For example, when a gate overdrive of up to 6V is applied (V0...), gs -V th At this time, the reported on-current density is approximately 0.5 mA / µm (at 100 nm L). ch (Reference 5). However, at very large-scale technology nodes, the power supply voltage (V) dd It is typically below 1V, which limits the V. gs The available voltage range is [not specified]. Simultaneously, increasing the CNT density to 500 CNT / µm and reducing the channel length to 10 nm also provides an on-state current density of 0.8 mA / µm (at approximately 3 V gate overdrive) (Reference 6). However, due to the strong inter-CNT shielding at high CNT densities, improving the conductivity per CNT also presents challenges. As a result, at the same channel length, the on-state conductance per CNT decreases to less than 2 µA / CNT, approximately 10% of that of a single-channel CNT FET (Reference 33). Furthermore, a subthreshold swing of approximately 500 mV per decibel occurs due to disruptive cross-CNTs and the diameter distribution at high CNT densities. Using 3D DNA nanogrooves minimizes the formation of cross-CNTs. Therefore, by exploring the correlation between CNT spacing and on-state conductance, optimized CNT spacing can balance the competing demands for higher CNT density and lower inter-CNT interactions. Together with the short-channel design, this will maximize the on-state conductance of the multi-channel CNTFET.

[0150] The International Semiconductor Technology Roadmap (Reference 1) recommends reducing the subthreshold swing to 60-80 mV per decibel. It is noteworthy that reducing the subthreshold swing does not decrease the on-state conductance. A subthreshold swing of 60 mV per decibel has been reported in CNT FETs constructed from thin-film CNT arrays (Reference 34). However, the on-state current density is only 100 nA / µm, which is insufficient for high-performance electronic devices. According to our demonstration, the subthreshold swing of a multi-channel CNT FET is slightly higher than that of a single-channel CNT FET. Due to the absence of intersecting CNTs, the difference in diameter distribution is small (17 mV per decibel). Therefore, when CNTs with uniform diameter are available, 3D DNA nanogrooves can, in principle, construct multi-channel CNT FETs with the same subthreshold swing as single-channel CNT FETs. Further reducing the subthreshold swing to the hot electron limit (60 mV per decibel) or even lower depends on the gate efficiency. For example, using graphene contact designs, single-channel CNT FETs have been shown to have subthreshold swings of less than 60 mV per decibel and on-state currents of 8 μA / CNT (Reference 35). Integrating graphene contact designs into multi-channel CNT FETs can facilitate on / off switching compared to current metal contacts.

[0151] Higher CNT purity is also essential for improving the success rate of FET construction. For the planned CNT FET architecture, 95% semiconductor CNT purity yields a 73% success rate in a six-channel CNT FET and a 54% success rate in a twelve-channel FET. Considering that high-performance microprocessors contain up to one billion FETs, semiconductor CNT purity must exceed 99.99999998% to ensure all FETs function correctly.

[0152] To illustrate Example 2, a CNT array with a designer width and inter-array spacing is fabricated.

[0153] In digital circuits, it is common to have larger spacing values ​​outside individual FETs than the semiconductor channel pitch. For example, in silicon circuits, Samsung's 14nm technology node has a uniform fin pitch of 49nm (FET width less than 250nm); while the spacing between the two closest fins in adjacent FETs can be as high as 700nm, 13 times larger than the fin pitch. Similar spacing differences have been observed in Intel's 22nm, 14nm, and 10nm silicon technology nodes. The larger spacing between the two closest FETs can accommodate interconnect metal lines. Larger FET spacing can be tailored to different circuit architectures.

[0154] Existing thin-film methods employ post-assembly etching to fabricate arrays with designer widths, inter-array spacing, and CNT counts exceeding centimeter levels. A continuous CNT film is first laid, covering the entire surface of the substrate. Then, post-assembly etching (via oxygen plasma) is introduced to etch the CNTs away from the channel regions. Figure 35 A). Therefore, custom array widths and inter-array spacings can be fabricated based on the FET / circuit layout. Importantly, the inter-array spacing is necessary to avoid stray conduction paths (Reference 37) and to accommodate metal contacts. It has been reported that the presence of CNTs beneath the contacts reduces the adhesion of the metal contacts to the substrate surface (Reference 6). After post-assembly etching, the full-surface-covering CNT film is etched into several individual arrays with widths ranging from approximately 50 nm to several hundred nanometers to accommodate the FET layout.

[0155] In contrast, we demonstrate a different strategy for implementing the designer width, array spacing, and CNT count specified in the specification. Figure 35 (B) Using 3D DNA nanogrooves, CNT arrays, along with the designer's CNT spacing and CNT count, are assembled onto a fixed-width 3D DNA template. The CNT count for each array can be programmed using different template widths. The assembled CNT arrays are then arranged within a pre-formed PMMA cavity, followed by PMMA stripping and DNA removal. Without post-assembly etching, the specified inter-array spacing is displayed after centimeter-level orientation. Since the inter-array spacing is defined by photolithography of the PMMA cavity, it can, in principle, be further reduced to below 200 nm. Therefore, the maximum array density is approximately 10⁵ / cm, close to that of silicon fins at the 10 nm technology node (less than 3 × 10⁵ / cm). The array width and inter-array spacing of our method are also similar to those fabricated by post-assembly etching methods.

[0156] References for Example 2

[0157] 1.The International Technology Roadmap for Semiconductors:2013.

[0158] 2.G.Hills et al., Understanding energy efficiency benefits of carbonnanotube field-effect transistors for digital VLSI, IEEE Trans.Nanotech.17,1259-1269(2018).

[0159] 3.Q.Cao et al.,Arrays of single-walled carbon nanotubes with fullsurface overage for high-performance electronics,Nature Nanotech.8,180-186(2013).

[0160] 4.G.J.Brady,K.R.Jinkins,M.S.Arnold,Channel length scaling behavior intransistors based on individual versus dense arrays of carbon nanotubes,J.Appl.Phys.122,124506(2017).

[0161] 5.G.J.Brady et al.,Quasi-ballistic carbon nanotube array transistorswith current density exceeding Si and GaAs,Sci.Adv.2,e1601240(2016).

[0162] 6.Q.Cao,J.Tersoff,D.B.Farmer,Y.Zhu,S.-J.Han,Carbon nanotubetransistors scaled to a 40-nanometer footprint,Science 356,1369-1372(2017).

[0163] 7.H.A.Becerril,A.T.Woolley,DNA-templated nanofabrication,Chem.Soc.Rev.38,329-337(2009).

[0164] 8.M.R.Jones,N.C.Seeman,C.A.Mirkin,Programmable materials and thenature of the DNA bond,Science 347,1260901(2015).

[0165] 9.S.-P.Han,H.T.Maune,R.D.Barish,M.Bockrath,W.A.Goddard,III,DNA-linker-induced surface assembly of ultra dense parallel single walled carbonnanotube arrays,Nano Lett.12,1129-1135(2012).

[0166] 10.A.Mangalum,M.Rahman,M.L.Norton,Site-specific immobilization ofsingle-walled carbon nanotubes onto single and one-dimensional DNA origami,J.Am.Chem.Soc.135,2451-2454(2013).

[0167] 11.H.Pei et al.,Organizing end-site-specific SWCNTs in specific lociusing DNA,J.Am.Chem.Soc.141,11923-11928(2019).

[0168] 12.W.Sun et al.,Precise pitch-scaling of carbon nanotube arrayswithin threedimensional DNA nanotrenches,Science(2020).

[0169] 13.H.T.Maune et al.,Self-assembly of carbon nanotubes intotwodimensional geometries using DNA origami templates,Nature Nanotech.5,61-66(2010).

[0170] 14.K.Keren,R.S.Berman,E.Buchstab,U.Sivan,E.Braun,DNA-templated carbonnanotube field-effect transistor,Science 302,1380-1382(2003).

[0171] 15.S.-J.Han et al.,High-speed logic integrated circuits withsolutionprocessed self-assembled carbon nanotubes,Nature Nanotech.12,861-865(2017).

[0172] 16.D.Zhong et al.,Gigahertz integrated circuits based on carbonnanotube films,Nature Electron.1,40-45(2018).

[0173] 17.C.Zhao et al.,Exploring the performance limit of carbon nanotubenetwork film field-effect transistors for digital integrated circuitapplications,Adv.Funct.Mater.29,1808574(2019).

[0174] 18.L.Liu et al.,Carbon nanotube complementary gigahertz integratedcircuits and their applications on wireless sensor interface systems,ACS Nano13,2526-2535(2019).

[0175] 19.A.M.Hung et al.,Large-area spatially ordered arrays of goldnanoparticles directed by lithographically confined DNA origami,NatureNanotech.5,121-126(2010).

[0176] 20.N.Nakatsuka et al.,Aptamer-field-effect transistors overcome Debyelength limitations for small-molecule sensing,Science 362,319-324(2018).

[0177] 22.F.Zhou et al.,Programmably shaped carbon nanostructure fromshapeconserving carbonization of DNA,ACS Nano 10,3069-3077(2016).

[0178] 23.Q.Cao,S.-J.Han,G.S.Tulevski,Fringing-field dielectrophoreticassembly of ultrahigh-density semiconducting nanotube arrays with a self-limited pitch,Nature Commun.5,5071(2014).

[0179] 24.J.Wu et al.,Self-assembly of semiconducting single-walled carbonnanotubes into dense,aligned rafts,Small 9,4142-4148(2013).

[0180] 25.Y.Cao et al.,Radio frequency transistors using alignedsemiconducting carbon nanotubes with current-gain cutoff frequency andmaximum oscillation frequency simultaneously greater than 70 GHz,ACS Nano 10,6782-6790(2016).

[0181] 26.J.Wu,A.Antaris,M.Gong,H.Dai,Top-down patterning and selfassemblyfor regular arrays of semiconducting single-walled carbon nanotubes,Adv.Mater.26,6151-6156(2014).

[0182] 27.D.Lee et al.,Three-dimensional fin-structured semiconductingcarbon nanotube network transistor,ACS Nano 10,10894-10900(2016).

[0183] 28.M.M.Shulaker et al.,High-performance carbon nanotube field-effecttransistors,IEEE Int.Electron Devices Meeting in San Francisco CA 33.6.1-33.6.4(2014).

[0184] 29.Y.Hu et al.,Growth of high-density horizontally aligned SWNTarrays using Trojan catalysts,Nature Commun.6,6099(2015).

[0185] 30.L.Zhang et al.,Assessment of chemically separated carbon nanotubesfor nanoelectronics,J.Am.Chem.Soc.130,2686-2691(2008).

[0186] 31.L.Zhang et al.,Optical characterizations and electronic devices ofnearly pure(10,5)single-walled carbon nanotubes,J.Am.Chem.Soc.131,2454-2455(2009).

[0187] 32.A.Gopinath,P.W.K.Rothemund,Optimized assembly and covalentcoupling of single-molecule DNA origami nanoarrays,ACS Nano 8,12030-12040(2014).

[0188] 33.C.Qiu et al.,Scaling carbon nanotube complementary transistors to5-nm gate lengths,Science 355,271-276(2017).

[0189] 34.L.Xiang et al.,Low-power carbon nanotube-based integrated circuitsthat can be transferred to biological surfaces,Nature Electron.1,237-245(2018).

[0190] 35.C.Qiu et al.,Dirac-source field-effect transistors as energy-efficient,highperformance electronic switches,Science 361,387-392(2018).

[0191] 36.G.S.Tulevski et al.,Toward high-performance digital logictechnology with carbon nanotubes,ACS Nano 8,8730-8745(2014).

[0192] 37.J.Zhang et al.,Robust digital VLSI using carbon nanotubes,IEEETransactions on Computer-Aided Design of Integrated Circuits and Systems 31,453-471(2012).

[0193] It should be noted that although several steps of the method for depositing nanostructures on a substrate have been described above, this division is merely exemplary and not mandatory. In practice, according to embodiments of this application, the features and functions of the two or more modules described above can be embodied in a single module. Conversely, the features and functions of a single module described above can be further divided into multiple modules.

[0194] By studying the specification, this application, the drawings, and the appended claims, those skilled in the art will be able to understand and implement other variations of the disclosed embodiments. In the claims, the terms "comprising" or "including" do not exclude other elements and steps, and the terms "a" or "an" do not exclude a plurality. In practical application of this application, a portion may perform the functions of multiple technical features referenced in the claims. No reference numerals in the claims should be construed as limiting the scope.

Claims

1. A method for depositing nanostructures on a substrate, characterized in that, The method comprises: forming a patterned alignment layer on a surface of the substrate, wherein the patterned alignment layer has one or more cavities, each cavity having a body region for containing at least one nanostructure template therein and a plurality of extension regions extending from and in fluid communication with the body region, wherein the plurality of extension regions are sized and shaped to not contain the at least one nanostructure template; and diffusing a nanostructure template into the one or more cavities of the patterned alignment layer.

2. The method of claim 1, wherein, The method comprises: removing the patterned alignment layer from the substrate.

3. The method of claim 2, wherein, Removing the patterned alignment layer from the substrate comprises: removing the patterned alignment layer from the substrate using a lift-off process.

4. The method according to any one of claims 1 to 3, characterized in that, The body region of each cavity of the one or more cavities is further configured to orient the at least one nanostructure template contained therein.

5. The method according to any one of claims 1 to 3, characterized in that, Each cavity of the one or more cavities is further configured to restrict at least one translational degree of freedom of the at least one nanostructure template along the surface of the substrate when the at least one nanostructure template is contained in the cavity.

6. The method of claim 1, wherein, Forming a patterned alignment layer on a surface of the substrate comprises: forming an alignment layer on a surface of the substrate; and patterning the alignment layer.

7. The method of claim 6, wherein, Patterning the alignment layer comprises: patterning the alignment layer using a photolithography process.

8. The method of claim 7, wherein, The photolithography process comprises: electron beam lithography, optical lithography, imprint lithography, or directed self-assembly of block copolymers.

9. The method of claim 7, wherein, Forming an alignment layer on a surface of the substrate comprises: spin coating a photoresist on a surface of the substrate to form the alignment layer.

10. The method of claim 9, wherein, The photoresist comprises: polymethyl methacrylate (PMMA) or polydimethylsiloxane (PDMS).

11. The method of claim 1, wherein, Diffusing a nanostructure template into the one or more cavities of the patterned alignment layer comprises: immersing a solution containing the nanostructure template on the patterned alignment layer; and incubating the substrate to diffuse the nanostructure template into the cavities.

12. The method of claim 11, wherein, Incubating the substrate comprises: dehydrating or evaporating the substrate in a sealed chamber for a predetermined period of time.

13. The method of claim 1, wherein, Prior to diffusing a nanostructure template into the one or more cavities of the patterned alignment layer, the method further comprises: differentiating the substrate to improve adhesion of the surface of the substrate to the nanostructure template.

14. The method of claim 2, wherein, The nanostructure template is decorated with a functional nanostructure, wherein, after removing the patterned alignment layer from the substrate, the method further comprises: removing a template portion of the nanostructure template from the substrate to leave the functional nanostructure on the substrate.

15. The method of claim 14, wherein, Prior to removing a template portion of the nanostructure template from the substrate, the method further comprises: forming a fixation structure on the substrate to fix the functional nanostructure on the substrate.

16. The method of claim 1, wherein, One of the one or more cavities has a plurality of extended regions that increase the volume of the cavity by at least 5%, 10%, 20%, 30%, 50%, 100%, 150%, or 200%.

17. The method of claim 1, wherein, Each of the one or more cavities has a depth that is 2 or more times the thickness of the nanostructure template.

18. The method of claim 1, wherein, The nanostructure template comprises one or more substances selected from the group consisting of nucleic acid templates, modified nucleic acid templates, protein templates, peptide nucleic acid templates, carbon nanotubes (CNTs), polymer- wrapped CNTs, CNT films, semiconductor nanoparticles, semiconductor nanowires, semiconductor nanoblocks, metal nanoparticles, metal nanowires, metal nanoblocks, polymeric nanoparticles, polymeric nanowires, polymeric nanoblocks, ceramic nanoparticles, ceramic nanowires, ceramic nanoblocks, metal oxide nanoparticles, metal oxide nanowires, metal oxide nanoblocks, fluoride nanoparticles, fluoride nanowires, and fluoride nanoblocks.

19. The method of claim 1, wherein, The nanostructure template comprises a polymer template.

20. The method of claim 1, wherein, The nanostructure template is modified with a functional nanostructure.

21. The method of claim 20, wherein, The functional nanostructure comprises one or more substances selected from the group consisting of carbon nanotubes (CNTs), polymer-wrapped CNTs, CNT films, semiconductor nanoparticles, semiconductor nanowires, semiconductor nanoblocks, metal nanoparticles, metal nanowires, metal nanoblocks, polymeric nanoparticles, polymeric nanowires, polymeric nanoblocks, ceramic nanoparticles, ceramic nanowires, ceramic nanoblocks, metal oxide nanoparticles, metal oxide nanowires, metal oxide nanoblocks, fluoride nanoparticles, fluoride nanowires, and fluoride nanoblocks.

22. A nanostructure array comprising at least one nanostructure template deposited on a substrate by the method of any one of claims 1 to 21.

23. A nanostructure array comprising at least one functional nanostructure deposited on a substrate by the method of any one of claims 14 to 15.

Images