KR20080050581A - Nano-molding process - Google Patents

Abstract

A nano-molding process including an imprint process that replicates features sizes less than 7 nanometers. The nano-molding process produces a line edge roughness of the replicated features that is less than 2 nanometers. The nano-molding process including the steps of: a) forming a first substrate having nano-scale features formed thereon, b) casting at least one polymer against the substrate, c) curing the at least one polymer forming a mold, d) removing the mold from the first substrate, e) providing a second substrate having a molding material applied thereon, f) pressing the mold against the second substrate allowing the molding material to conform to a shape of the mold, g) curing the molding material, and h) removing the mold from the second substrate having the cured molding material revealing a replica of the first substrate.

Description

Nano-molding process

The present invention relates to a nanoforming method.

New technologies for fabricating nanometer-sized structures are critical to advances in nanoscience and technology. As the demand for smaller electronic devices and biological assay devices increases, there is a need for improved manufacturing methods for manufacturing such devices. The method can be used in the manufacture of electronic, magnetic, mechanical and optical devices as well as biochemical analytical devices. The method can be used, for example, to define features and arrangements of microcircuits as well as the structure and operating features of optical waveguides and components.

These methods can also play an important role in the semiconductor industry, replacing projection mode photolithography, which, due to practical limitations, cannot reach resolutions below 45 nm. Projected photolithography is a method of patterning features in which a thin layer of photoresist is applied to a substrate surface and a selected portion of the resist is exposed to patterned light. The resist is then developed to reveal the desired pattern of the exposed substrate for further processing, such as etching. The problem with this method is that the resolution is limited by the wavelength of light, scattering in resists and substrates, and thickness and properties of resists. As a result, projection photolithography cannot be used to economically produce feature sizes of less than 100 nm.

Next-generation lithography (NGL), which includes electron beams, dip pens, and nano-imprint technologies, is being considered. Electron beam methods include the use of microlithography, which creates patterns in polymers, so-called resists, and is based on short-wave UV radiation or electron beams. The pattern is formed due to a change in solubility of the polymer by exposure to imaging radiation while using a solvent to remove a portion of the polymer film. However, the large scale generation of length scale dimensions below 100 nm using these techniques is costly and can be accomplished using very specialized imaging tools and materials.

Among the NGL techniques, the technique of using a mold for imprinting a feature on a polymer thin film has attracted considerable attention. Well-defined optics combined with photolithography techniques can accurately specify their resolution, but the resolution limits of NGL based nanoforming are much harder to determine. Uncertain polymer physics dominating the molding process and the lack of reliable means to assess resolutions less than 5 nm in length represent some limitations to the current technology.

Accordingly, there is a need in the art for a nanoforming method that can be used to solve the problems outlined above and to produce relief structures with lateral and vertical dimensions of less than 10 nm. There is also a need in the art for a method that enables identification of dimensions in parts made by the method.

Summary of the Invention

Nanoforming methods include an imprinting process that replicates feature sizes of less than 7 nm. Nanoforming methods produce line edge roughness of replication features that are less than 2 nm. The nanoforming method comprises a) forming a first substrate having nanoscale features thereon, b) casting one or more polymers onto the substrate, and c) curing the one or more polymers to form a mold. d) removing the mold from the first substrate, e) providing a second substrate with a molding material applied thereon, f) pressing the mold against the second substrate to conform the molding material to the shape of the mold. Steps, g) curing the molding material, and h) removing the mold from the second substrate having the cured molding material to reveal a replica of the first substrate.

1 graphically illustrates the nanoforming method of the present invention.

2 (a-d) are atomic micrographs (a) of a single wall carbon tube of a first substrate and three separate replicas (b-d) made by the method of the present invention.

FIG. 3 (ac) shows two atomic force micrographs of an atomic force micrograph (a) of a replica formed by the method of the present invention and a single-wall carbon tube (c) of the replica formed by the method of the present invention (b) and the first substrate. It includes.

Figure 4 (ah) is a graph of the height of a feature formed in a molding material bonded to individual carbon nanotubes with diameters of 2, 1.3 and 0.9 nm (a), the length of the feature for the replica as a function of single-wall carbon nanotube diameter. Graph of averaged heights (b), atomic force micrographs (ce) of individual carbon nanotubes with diameters 2, 1.3 and 0.9 and atomic force micrographs of replicas combined with individual carbon nanotubes with diameters 2, 1.3 and 0.9 ( fh).

Referring to Figure 1, the nanoforming method of the present invention is shown graphically. The method comprises the steps of a) forming a first substrate having nanoscale features formed thereon, b) casting one or more polymers to the first substrate, c) curing the one or more polymers to form a mold, d) Removing the mold from the first substrate, e) providing a second substrate with a molding material applied thereon, f) pressing the mold against the second substrate to conform the molding material to the shape of the mold, g ) Hardening the molding material and h) removing the mold from the second substrate having the cured molding material to reveal a replica of the first substrate.

In a preferred aspect of the present invention, the first substrate is a silicon wafer having a single-walled carbon nanotube (SWNT) formed thereon. The silicon wafer may include a silicon dioxide (SiO ₂ ) layer formed thereon to facilitate adhesion and formation of SWNTs. The first substrate may also include other materials such as electron beam lithography, molds made by x-ray lithography or biological materials on plastic sheets. The first substrate preferably comprises high quality sub-monolayers of small diameter SWNTs that serve as a template from which nanomoulds can be constructed. SWNT is a method of the invention because of the cylindrical cross-section and high aspect ratio of the tubes, atomic scale uniformity of their dimensions over multiple micron lengths, their chemical inertness and their ability to grow or deposit them in large quantities on the large area of the substrate Suitable for use on

SWNTs can be formed using methane-based chemical vapor deposition using relatively high concentrations of ferritin catalysts. The SWNT formed has a diameter of 0.5 to 10 nm, preferably 0.5 to 5 nm and may have an amount of 1 to 10 tubes / m ² applied to the SiO ₂ / Si wafer. Due to the diameter of the continuous range of tubes and their relatively high but monolayer application they are ideal for evaluating resolution or dimensional limits. The cylindrical geometry of SWNTs allows their dimensions to be characterized simply by atomic force microscopy (AFM) measurements of their heights. SWNTs are bonded to SiO ₂ / Si wafers by van der Waals bonding force, which bonds the SWNTs to the substrate with sufficient strength to prevent their removal when peeling away the cured polymer molds, Will be discussed in more detail below. Preferably, the absence of polymer residues in large areas of the SWNTs can replicate the fine resolution features formed in the first substrate. The lack of polymer residue indicates that the mold did not contaminate the master, so the features of the mold are by true replication and not by material breakage. Optionally, the SWNTs formed on the first substrate may include a silane layer that will act as a release agent applied thereto to prevent adhesion of the polymer used to form a mold of the first substrate.

After forming the first substrate, one or more polymers are cast and cured on the first substrate to form a mold. In a preferred aspect of the invention, the mold is a composite mold having a plurality of polymer layers. The first layer applied to the first substrate is a relatively high modulus (about 10 MPa) elastomer (h-PDMS) based on polydimethylsiloxane. h-PDMS is preferably prepared by mixing vinylmethylsiloxane-dimethylsiloxane, 1,3,5,7-tetravinyl-1,3,5,7-tetramethylcyclotetrasiloxane and platinum catalyst. Methylhydrosiloxane-dimethylsiloxane is then added and mixed to form a prepolymer mixture of h-PDMS. The prepolymer mixture of h-PDMS may be spin casted onto the first substrate or alternatively by casting the prepolymer mixture of h-PDMS onto the first substrate. The prepolymer mixture of h-PDMS is partially cured by a platinum catalyst, which adds SiH bonds across the vinyl group to the prepolymer mixture of h-PDMS to form SiCH ₂ -CH ₂ -Si bonds (hydrosilylation process Also known as Multiple reaction sites in both the base and the crosslinkable oligomer in the prepolymer mixture of h-PDMS allow 3D crosslinking to prevent relative migration between the bound atoms. The low viscosity of the prepolymer mixture of h-PDMS (about 1000 cP at room temperature) and the conformability of the silicone backbone allow for the replication of microfeatures.

The second polymer is applied to the back side of the partially cured h-PDMS layer. Preferably, the mold can be handled more easily by applying a physically coarse low modulus PDMS (s-PDMS) layer to the partially cured h-PDMS layer. In a preferred aspect of the invention, the s-PDMS is Sylgard 184, commercially available from Dow Corning Corporation. After applying the s-PDMS layer, the plurality of polymer layers on the first substrate can be sufficiently cured to form a composite mold.

After forming the mold, the molding material is applied to the second substrate and the mold is pressed onto the second substrate so that the molding material matches the shape of the mold. The second substrate is preferably an SiO ₂ / Si wafer as already described with respect to the first substrate. Other materials can also be used, including wafers, plastic films, glass plates, or other materials suitable for the purpose of acting as a substrate.

The molding material may be a prepolymer, a monomer or any polymer that can be molded using a composite mold and may have the properties required for analysis. Preferably, the molding material is a polyurethane prepolymer (PU), even more preferably a photocurable or ultraviolet curable PU polymer. In addition, the molding material may be polyacrylic acid (PAA) that can be analyzed by TEM imaging. PU formulations preferably comprise a prepolymer, a chain extender, a catalyst and an adhesion promoter. By lightly pressing the mold against the layer, the liquid PU prepolymer can flow and conform to the relief features on the mold. Passing and crosslinking the PU by passing light through a transparent mold can yield a cured PU having a Shore D hardness of 60. After curing the molding material, the mold is removed to reveal a replica of the features formed in the first substrate.

Direct AFM characterization of the PU surface reveals the vertical dimension of the imprinted relief with atomic scale precision. As outlined above, the method of the present invention may utilize other curing techniques, but curing is preferably photocuring.

2 shows an AFM picture of SWNT applied to a first substrate and corresponding regions of three different PU structures imprinted into a single mold derived from the first substrate or master. Qualitatively, the data show that the method of the present invention accurately reproduces nanoscale features associated with SWNTs even during multiple imprinting cycles. Both Y-type SWNT bonds and smaller tube fragments on the first substrate can be seen in each PU sample. The line scan collected from the lower left side chain of the "Y" structure (insert Figure 2) shows that the imprinted relief feature has a similar height as the relief feature on the master. Some of the obvious variations in the cross-sectional shape of these features can be attributed to the AFM products associated with the roughness of the molded PU surface. This roughness has an average square root amplitude (evaluated by AFM) of 0.37 nm for clone 1 and 0.4 nm for clone 2 and clone 3. The maximum peak to valley height change associated with this roughness is about 1.5 nm.

The photograph obtained with AFM accurately reveals only the height of the relief feature. Transmission electron microscopy (TEM) can measure their width. Since polyacrylic acid (PAA) is an established polymer for TEM analysis, PAA rather than PU was imprinted for this purpose. FIG. 3A shows AFM images evaluated in the same area of the PAA layer imprinted using the same PDMS mold used in the results of FIG. 2. Replication fidelity, feature height, surface roughness and other properties are similar to those observed in the PU. Pt / C was deposited (at 30 ° on the PAA surface) (to provide contrast in the TEM), then C was deposited on the imprinted PAA (at normal angle of incidence to provide structural support for Pt) and Dissolving the PAA creates a Pt / C membrane replica of the relief structure. For comparison, a similar Pt / C replica was prepared by etching the SiO ₂ layer (with 2% HF in H ₂ O) from the SWNT master to peel off the replica. 3 shows TEM photographs of two types of replicas. Dark and light stripes along the tube feature represent metal deposits and shadows, respectively. Separation between the darkest and lightest areas roughly limits the width of the feature. Separation was measured by analyzing the line scan of the image averaged over the straight length (50 nm) of the corresponding relief feature. The profiles measured in this way from the imprinted PAA and SWNT masters show similar shapes. In both cases a width range of about 3 nm to about 10 nm was observed. Widths below 3 nm are difficult to measure at least in part due to the visible granule size (about 1 nm) of Pt / C. The TEM data of the replica is consistent with the dimensions and cross-sectional shape of the master.

From the AFM and TEM photographs shown in FIGS. 2-4 it is clear that SWNTs with a diameter of more than 2.5 nm on the master clearly look like continuous replicated features in the replica. However, the height of these features varies along their length from a maximum value that is approximately equal to the height of the SWNT on the master to a minimum value that is equal to the minus the height equivalent to the peak-to-valley surface roughness. Roughness plays a role in polymer properties that limit resolution and is an indicator of polymer properties.

Roughness, including line edge roughness and peak-to-valley roughness, is due to clearly missed cross-sections or fracture occurrences that begin to appear in AFM images of relief features associated with SWNTs less than 2 nm in diameter, as shown in FIG. In a preferred aspect of the invention, the line edge roughness and surface roughness are less than 2 nm. In the case of 1-2 nm diameter SWNTs, this fracture represents a significant fraction of the total length of the imprinted structure. For less than 1 nm, only small fragments of the replica construct can be seen, as shown in FIG. 4A. However, even at about 1 nm scale, the replicated relief can still be identified by averaging the AFM lines collected along the length of the feature as shown in FIG. 4B. The resolution limit can be summarized by plotting the position averaged relief height as a function of the SWNT diameter as shown in FIGS. 4A-4E. Ultimate resolution correlates to the ability of the prepolymer (PDMS, PU or PAA) to conform to the surface (master or PDMS mold) and the ability of the polymer (PDMS, PU or PAA) to maintain the molded shape. Three parts of data suggest that the PDMS mold defines the resolution. First, in the molded relief feature, the fractures shown in FIGS. 4F-4H occur at the same location in multiple molding cycles. Second, structures imprinted on dissimilar polymers (ie PU and PAA) have similar surface roughness and relief height distributions. Third, the PU formed on the exposed fluorinated SiO ₂ / Si wafer produces surface roughness (0.19 nm), which is smaller than that produced using a flat PDMS mold derived from the same wafer.

For extremely high resolution features, there are two or more important length scales that affect the resolution limit of the replica: (i) for h-PDMS, the average distance between crosslinks, on the order of about 1 nm, and (ii) chemistry in the 0.2 nm range. Combined length. The correlation between the resolution limit and the length scale for features using different PDMS polymers is shown in Table 1. While it is difficult to assign observed roughness and resolution limits to specific molecular features of PDMS, crosslink density seems to be an important parameter. The average molecular weight between crosslinks (Mc) and the distance between crosslinks (D) were measured by swelling the sample in toluene and applying the Flory-Huggins theory. Table 1 summarizes the experimental resolution limits and roughness parameters for Mc and D values and h-PDMS, and low crosslink density modalities of the material (hl-PDMS) and commercially available low modulus PDMS (s-PDMS). These three materials show a qualitative correlation between resolution and crosslink density. They are also related to resolution and roughness, both of which are affected by the conformability of the polymer chains and the ability to retain the shape of the crosslinked polymer. Attempts to improve resolution by increasing the number of crosslinks in h-PDMS have failed due to the tendency of the resulting material to stick to the SWNT first substrate.

Crosslinking distance theory (nm) Mc theory (g / mol) Mc experimental value (g / mol) rms roughness (nm) Peak-to-Valley Roughness (nm) Resolution limit approximation h-PDMS 1.28 357 377 0.37 1.7 2 hl-PDMS 1.6 554 536 0.54 3.1 3 s-PDMS 2.7 1239 891 0.54 3.2 3.5

Using the following example procedure to prepare a nano-molded product analyzed and shown in Figures 2 to 4.

Preparation of Carbon Nanotube Master: A silicon wafer having a SiO ₂ layer (thermal growth) having a thickness of 100 nm was provided as a substrate for SWNT growth. A ferritin catalyst (Aldrich) diluted with deionized water at a capacity ratio of 1: 1000 was cast onto the wafer. The wafer was then immediately placed in a quartz tube at 800 ° C. for 2 minutes and then purged with hydrogen gas at 900 ° C. for 1 minute. Methane [500 standard minute flow rate (cm ³ / min, sccm)] and hydrogen (75 sccm) were flowed into a quartz tube at 900 ° C. for 10 minutes to grow SWNTs.

Mold formation: SWNT / SiO ₂ / Si master with 100 μl of (tridecafluoro-1,1,2,2-tetrahydrooctyl) -1-trichlorosilane (United Chemical Tech) in a vacuum chamber for 2 hours Put in. The resulting silane layer (a monolayer or submonoscopic application is expected) prevents PDMS from adhering to the exposed SiO ₂ . h-PDMS (Gelest, Inc) was prepared as follows: 3.4 g (7-8% vinylmethylsiloxane) (dimethylsiloxane), (1,3,5,7-tetravinyl-l, 3,5,7 100 µg of tetramethylcyclotetrasiloxane) and 50 µg of platinum catalyst were mixed and placed in a vacuum chamber for 5 minutes. Subsequently, 1 g of (25-30% methylhydrosiloxane) (dimethylsiloxane) was added and mixed, and the resulting sample was put back into the vacuum for 5 minutes. The prepolymer mixture was spin casted at 1000 rpm for 40 seconds on a SWNT master and then baked at 65 ° C. for 4 minutes. S-PDMS (Silgard 184, Dow Corning) prepared by mixing the substrate and curing agent in a ratio of 10: 1 was poured onto h-PDMS. Baking at 65 ° C. for 2 hours to complete curing of the polymer. Typical thicknesses were 10 μm for h-PDMS and 3 mm for s-PDMS.

Molded Material: Polyurethane (PU) (NOA 73, Norland Products) was spin casted onto a SiO ₂ / Si wafer at 9000 rpm for 40 seconds. The mold was placed on the thin film and gently pressed to ensure good wetting at the interface. The PU was exposed to ultraviolet (350-380 nm; long wavelength UV lamp, UVP) through the mold at about 19mw / cm ² for 1 hour to cure the PU and solidify the film.

Measurement Analysis: Characterization of SWNT master and imprinted PU structures was performed with AFM (Dimension 3100, Digital Instrument) and TEM (Philips CM200, FEI). TEM photographs of metal blocked replicas were taken at 120 kV. AFM measurements were performed in tapping mode using a tip supplied by BudgetSensors (BS-Tap300Al). The resonant frequency of the tip was 30 kHz.

Preparation of TEM Samples: A few drops of methanol solution of polyacrylic acid (PAA) (30 wt.%) Was dropped onto a PDMS mold. Samples were left in this configuration in open air until methanol evaporated (typically about 10 hours was sufficient). The PAA film was then peeled off to leave the imprinted nanostructures on the PAA surface. The sample was placed in a vacuum chamber of a thermal evaporator. The Pt / C source was placed on the sample at an elevation angle of 30 °. Several nm of Pt / C was deposited on the sample. The carbon film, about 10 nm thick, was then evaporated at normal angle of incidence with respect to the sample. Samples were immersed in deionized water for several hours until PAA dissolved. Subsequently, Pt / C and the carbon film were drifted on the water surface. These were later collected by TEM copper mesh.

The invention has been described in an illustrative manner and the terminology used is to be understood as illustrative and not restrictive.

Many modifications and variations of the present invention are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims, the invention may be practiced otherwise than as specifically described.

Claims (19)

A nanoforming method comprising an imprinting process capable of producing a replica having a line edge roughness of less than 2 nm. The method of claim 1, wherein the imprinting process is capable of replicating one or more features having a minimum lateral dimension of less than 7 nm. The method of claim 1, wherein the intrinsic surface roughness measured by the peak-to-valley surface roughness of the material is less than 2 nm. The nanoforming method of claim 2, wherein the intrinsic surface roughness measured by the peak-to-valley surface roughness of the material is less than 2 nm. The method of claim 1, a) forming a first substrate having nanoscale features formed thereon; b) casting at least one polymer on the first substrate; c) curing one or more polymers to form a mold; d) removing the mold from the first substrate; e) providing a second substrate coated with a molding material thereon; f) pressing the mold with a second substrate to conform the molding material to the shape of the mold; g) curing the molding material and h) removing the mold from the second substrate with the cured molding material to reveal a replica of the first substrate. The method of claim 5, wherein the step of casting and curing one or more polymers a) casting the first polymer to the first substrate; b) partially curing the first polymer; c) applying a second polymer to the first polymer and d) curing the first polymer and the second polymer to form a composite mold. The method of claim 6, wherein the first polymer comprises h-polydimethylsiloxane. The method of claim 6, wherein the second polymer comprises s-polydimethylsiloxane. The method of claim 5, wherein the first substrate has nanoscale features with dimensions greater than 2 nm. The nanoforming method of claim 9, wherein the dimension is between 2 and 80 nm. The method of claim 9, wherein the dimension is 2 to 7 nm. The method of claim 5, wherein forming the first substrate comprises forming single-walled carbon nanotubes on the silicon wafer. The nanoforming method of claim 5, wherein the molding material is a photocurable material. The method of claim 13, wherein the photocurable material is selected from the group consisting of polyurethane or vinyl functional monomers. The method of claim 5, further comprising the step of identifying the dimensions of the replica. The method of claim 15, wherein the verifying step a) measuring a vertical dimension with respect to the first substrate; b) measuring the vertical dimension for the replica and c) comparing the vertical measurements of the first substrate and the replica. The method of claim 16, wherein the vertical dimension is measured using an atomic force microscope. The method of claim 15, wherein the verifying step a) measuring lateral dimensions for the first substrate; b) measuring the side dimensions of the replica and c) comparing the side measurements of the replica with the first substrate. The method of claim 18, wherein the lateral dimensions are measured using transmission electron microscopy.

Images