Classifications

• • G—PHYSICS
  • G03—PHOTOGRAPHY; CINEMATOGRAPHY; ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ELECTROGRAPHY; HOLOGRAPHY
  • G03F—PHOTOMECHANICAL PRODUCTION OF TEXTURED OR PATTERNED SURFACES, e.g. FOR PRINTING, FOR PROCESSING OF SEMICONDUCTOR DEVICES; MATERIALS THEREFOR; ORIGINALS THEREFOR; APPARATUS SPECIALLY ADAPTED THEREFOR
  • G03F7/00—Photomechanical, e.g. photolithographic, production of textured or patterned surfaces, e.g. printing surfaces; Materials therefor, e.g. comprising photoresists; Apparatus specially adapted therefor
  • G03F7/0002—Lithographic processes using patterning methods other than those involving the exposure to radiation, e.g. by stamping
• • B—PERFORMING OPERATIONS; TRANSPORTING
  • B82—NANOTECHNOLOGY
  • B82Y—SPECIFIC USES OR APPLICATIONS OF NANOSTRUCTURES; MEASUREMENT OR ANALYSIS OF NANOSTRUCTURES; MANUFACTURE OR TREATMENT OF NANOSTRUCTURES
  • B82Y40/00—Manufacture or treatment of nanostructures
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01Q—SCANNING-PROBE TECHNIQUES OR APPARATUS; APPLICATIONS OF SCANNING-PROBE TECHNIQUES, e.g. SCANNING PROBE MICROSCOPY [SPM]
  • G01Q60/00—Particular types of SPM [Scanning Probe Microscopy] or microscopes; Essential components thereof
  • G01Q60/24—AFM [Atomic Force Microscopy] or apparatus therefor, e.g. AFM probes

Definitions

• the present invention relates generally to nanoimprint lithography, and more particularly to developing wafer-scale programmable films for semiconductor planarization and for imprint lithography.
• Nanoimprint lithography is a method of fabricating nanometer scale patterns that is low cost, high throughput and high resolution. It creates patterns by mechanical deformation of imprint resist and subsequent processes.
• the imprint resist is typically a monomer or polymer formulation that is cured by heat or UV light during the imprinting. Adhesion between the resist and the template is controlled to allow proper release.
• J-FIL Jet and Flash Imprint Lithography
• a method for fabricating patterns on rigid wafer substrates comprises implementing an inverse optimization scheme to determine process parameters used to obtain a desired film thickness of a liquid resist formulation, where the liquid resist formulation comprises a solvent and one or more non-solvent components.
• the method further comprises covering a substrate with a substantially continuous film of the liquid resist formulation using spin-coating, where the liquid resist formulation is diluted in the solvent.
• the method additionally comprises evaporating the solvent substantially from the liquid resist formulation forming a film.
• the method comprises closing a gap between a template and the substrate.
• the method comprises curing the film to polymerize the film.
• the method comprises separating the substrate from the template leaving the polymerized film on the substrate.
• a method for planarizing patterns on rigid wafer substrates comprises implementing an inverse optimization scheme with existing substrate topography as an input which is used to determine process parameters used to obtain a desired film thickness of a liquid resist formulation, where the liquid resist formulation comprises a solvent and one or more non-solvent components.
• the method further comprises covering a patterned substrate with a substantially continuous film of the liquid resist formulation using spin-coating, where the liquid resist formulation is diluted in the solvent.
• the method additionally comprises evaporating the solvent substantially from the liquid resist formulation forming a film.
• the method comprises curing the film to polymerize the film.
• a method for performing in-situ metrology comprises having an array of Atomic Force Microscope (AFM) tips with individually addressable locations.
• the method further comprises determining an optimum positioning of the array of AFM tips on a wafer with a pattern, a topography variation or a variation in a film thickness, wherein each AFM tip in the array of AFM tips provides data that is aggregated to infer the pattern, the topography variation or the film thickness variation.
• the method additionally comprises scanning the wafer to measure the pattern, the topography variation or the film thickness variation.
• the method comprises aggregating and analyzing the measured pattern, the measured topography variation or the measured film thickness variation to infer the pattern, the topography variation or the film thickness variation.
• Figure 1 is a flowchart of a method of the inkjet-based P-FIL (IJ-PFIL) process for fabricating a pattern in accordance with an embodiment of the present invention
• Figures 2A-2F depict the cross-sectional views of fabricating a pattern using the steps described in Figure 1 in accordance with an embodiment of the present invention
• Figure 3 is a flowchart of method for fabricating replica templates and final substrates using roll-to-plate lithography in accordance with an embodiment of the present invention
• Figures 4A-4J depict the cross-sectional views of fabricating replica templates and final substrates using the steps described in Figure 3 in accordance with an embodiment of the present invention
• Figure 5 is a diagram of the framework for the inverse optimization scheme for D-PFIL in accordance with an embodiment of the present invention.
• Figure 6 is a flowchart of a method of the spin-coated based P-FIL (SC-PFIL) process for fabricating a pattern in accordance with an embodiment of the present invention
• Figures 7A-7F depict the cross-sectional views of fabricating a pattern using the steps described in Figure 6 in accordance with an embodiment of the present invention
• Figure 8 is a flowchart of a method of an alternative embodiment of the spin-coated based P-FIL (SC-PFIL) process for fabricating a pattern in accordance with an embodiment of the present invention
• Figures 9A-9F depict the cross-sectional views of fabricating a pattern using the steps described in Figure 8 in accordance with an embodiment of the present invention
• Figure 10 is a diagram of the framework for the inverse optimization scheme for SC- PFIL in accordance with an embodiment of the present invention.
• Figure 11 is a diagram of the framework for the inverse optimization scheme for Slot Die (SD)-PFIL in accordance with an embodiment of the present invention
• Figure 12 illustrates a stable bead formation in accordance with an embodiment of the present invention
• Figure 13 illustrates variation of film thicknesses with spin time and spin speed in accordance with an embodiment of the present invention
• Figure 14 is a flowchart of a method for fabricating nanoscale patterns made of high- index dielectric materials, such as Ti0 2 , on a flat surface in accordance with an embodiment of the present invention.
• Figures 15A-15D depict the cross-sectional views of fabricating nanoscale patterns on a flat surface using the steps described in Figure 14 in accordance with an embodiment of the present invention.
• P-FIL Programmable Film Imprint Lithography
• This process is compatible with various modes of material deposition, including inkjetting, slot die coating, and spin- coating, for coating a film of resist on rigid substrates.
• the process may also be used in conjunction with film coating processes, such as dip coating on substrates that are not substantially flat.
• the thickness of the resist film, just before the patterning step is substantially continuous and may be non-uniform depending on the given pattern density variations.
• This non-uniformity may be (i) substantially correlated with an initial non-uniform material distribution or, (ii) can be enhanced from an initial uniform distribution of material through controlled external inputs, such as distributed thermal actuation or controlled gas flow, to evaporate the resist film with spatial selectivity or, (iii) a combination of approaches (i) and (ii).
• distributed inkjetting of material can be used to add more material selectively on top of films that are deposited using spin coating and slot die coating, etc. This may be necessitated in cases where simultaneous patterning of micro- and nano-scale structures is desired.
• the thickness of the deposited film may be made compatible with the nano-scale structures, while inkjetting of larger drops may be used to cater to the micro-scale structures.
• the achievement of a programmed non-uniform and substantially continuous film is driven by an optimization program that informs the initial material distribution and the nature of external inputs, such as evaporation or gas flow.
• the gap between the resist-coated substrate and a patterned template is closed to fill the template pattern.
• the resist film is then cross-linked using UV light and the template is then separated from the substrate to leave the polymerized resist film on the substrate.
• an intentionally non-uniform film is formed on a pre-patterned substrate or an un-patterned substrate using the methods described in the previous paragraph. This is done to compensate for substrate topography variations to achieve a desired final substrate topography.
• Exemplar desired final substrate topography may be one of the following: (i) flat or substantially flat as in the field of wafer polishing or fabrication of optical flats, or (ii) locally flat (over micrometer or millimeter length scales) while globally (over tens of millimeters or more) conforming to the incoming substrate topography as in the field of semiconductor planarization.
• a goal of the process of the present invention is to deposit an intentionally non-uniform film on a substrate where the resulting top surface has no patterns.
• the liquid resist material may be cross-linked in the free surface state or contact may be made against a superstate that is substantially free from nanoscale and microscale patterns.
• exemplar materials can include the use of low-viscosity materials, such as the one developed for reverse-tone step and flash imprint lithography, such as discussed in Lin, M.W., "Simulation and design of planarizing materials and interfacial adhesion studies for step and flash imprint lithography," Ph.D. Dissertation, The University of Texas at Austin, 2008, which is incorporated by reference herein in its entirety.
• the superstrate influences the evolution dynamics of the thin film. Its geometry can thus be optimized, by locally changing its thickness, to match the desired substrate topography and process specifications.
• resist drops that are diluted by solvents. Details of material design are discussed further below. This increases their volume without increasing the monomer content and can thus lead to favorable drop spreading and merging characteristics without using the template to force the formation of a contiguous film in an ambient environment that may not need process gases, such as carbon dioxide, Helium or PFP discussed earlier.
• Figure 1 is a flowchart of a method 100 of the inkjet-based P-FIL (IJ-PFIL) process for fabricating a pattern in accordance with an embodiment of the present invention.
• Figures 2A-2F depict the cross-sectional views of fabricating a pattern using the steps described in Figure 1 in accordance with an embodiment of the present invention.
• step 101 drops of a diluted monomer 201 (diluted in a solvent) are dispensed on a substrate 202 as shown in Figure 2A, which are spontaneously merged with one another as shown in Figure 2B, forming a liquid resist formulation 203 (containing the solvent and one or more non-solvent components, such as the monomer) as shown in Figure 2C.
• the inkjet is used to dispense discrete drops of diluted monomer which does not form a substantially continuous film to locally modify film thickness profiles prior to closing the gap between template 205 and substrate 202 (discussed below).
• the inkjet drop dispense and spontaneous drop merging stages can be done at a tool location where the temperature and air flow are highly controlled and kept intentionally low to minimize evaporation. This would be beneficial for inkjet drop dispensing as the material formulation would be more stable. Evaporation of material from the drop as it is in-flight or sitting on substrate 202 before and during merging is not desirable. As soon as the drops have merged to form a contiguous film, the processing can then be transferred to a tool location where the temperature is at a higher mean value than the location where drop merging happens. This higher mean temperature can take place through a heated substrate and or chuck, for example, and should make the solvent volatile while also favorably changing the surface tension and viscosity of the solution. It is also important to note here that the entire solvent volume needs to be evaporated, to prevent recondensation of the solvent on the substrate. Any remaining solvent can lead to voids or cracks in the cured resist, which is highly undesirable.
• step 102 solvent is evaporated (e.g., blanket evaporation) from liquid resist formulation 203 followed by selective multi-component resist film evaporation resulting in a non-uniform and substantially continuous film 204 on substrate 202 as shown in Figure 2D.
• the liquid resist formulation consists of solvent with a substantially higher volatility than the rest of the components of the liquid resist formulation.
• the non-solvent component(s) of the liquid resist formulation have substantially similar volatility.
• the non-solvent component(s) of the liquid resist formulation except for a photoinitiator have a boiling point that differ by less than 30° C, 10° C or 5° C.
• the non-solvent component(s) include a thermoset polymer with a high glass transition temperature.
• the non-solvent component(s) include silicon or carbon.
• an inverse optimization scheme may be used to determine process parameters used to obtain a desired film thickness of the liquid resist formulation.
• the inverse optimization scheme includes one or more of the following: genetic algorithms, simulated annealing, and full pattern search to achieve an objective function of minimizing an error between a function of the desired film thickness and the obtained film thickness.
• a forward model is used as a core for the inverse optimization scheme to predict a film thickness profile, where the forward model incorporates one or more of the following: thermal transport, solute transport, and solvent transport.
• the liquid resist formulation, whose solvent has already been evaporated is selectively evaporated to form film 204 with an intentionally non-uniform profile to match a desired thickness profile.
• the selective evaporation is carried out by one or more of the following: a temperature controlled substrate chuck, an array of individually controllable infrared lamps, an array of individually controllable fans to control air flow, individually controllable micro heaters or coolers on substrate 202, and a spatial light modulator (may include one or more Digital Micromirror Device (DMD) arrays) to control one or more sources of electromagnetic radiation.
• a spatial light modulator may include one or more Digital Micromirror Device (DMD) arrays
• DMD Digital Micromirror Device
• the solvent evaporation, selective resist evaporation and imprinting are carried out in stations that may be separated by barriers that prevent vapor transport between the stations.
• the selective evaporation is carried out in an environment with controlled vapor pressures.
• the control of resist film evaporation can be done in a separate station, which is physically separated with barriers, through judicious control of temperature through a distributed set of thermal actuators, infrared (IR) lamps, or combinations thereof, in which each element in the set can be individually controlled.
• This individual control in infra-red (IR) lamps can be achieved through devices, such as Digital Micromirror Device (DMD) Arrays from Texas Instruments.
• DMD Digital Micromirror Device
• a state-of-the-art array has 4M pixels at a pitch of 7.6 micrometers at a diagonal width of 0.9 inches. This can enable spatial modulation of IR intensity to control the evaporation over at least an area of - 0.4 sq. inch.
• a resist formulation containing methyl methacrylate (MMA) as the liquid monomer and primary component would need a local heat source of ⁇ 1 W to absorb sufficient heat at normal temperature and pressure to evaporate -100 nm of liquid film from a local 50 x 50 ⁇ 2 area, assuming the starting film thickness is 500 nm.
• Such powers can be delivered using DMD arrays. These devices also have frame rates approaching 10 kHz, and thereby provide high frequency temporal control of the incident radiation and subsequent evaporation.
• multiple DMD arrays can be used across a larger area, if resolution cannot be compromised further.
• the individual pixels may have significant cross-talk, which can also be modeled and incorporated in the forward model for thin film evaporation.
• an underlying substrate with MEMS-based micro-heaters may also be used for enhanced temperature distribution, if a higher resolution is required than what can be afforded by the DMD arrays. This principle of selective heating can be generally applied to all embodiments of the present invention and is not specific to only IJ-PFIL.
• template 205 is patterned with a lateral feature dimension of less than 500 nm. In one embodiment, template 205 is patterned with a lateral feature dimension of less than 100 nm. In one embodiment, the template consists of a three-dimensional shape or a multi-tiered pattern as discussed in P.
• contact between the substrate and the template is initiated substantially along a line or substantially at a point. This is done to prevent substantial area contact, as it can potentially lead to trapping of bubbles and voids because of the flexible nature of the substrate.
• precise motion control and/or force control is employed while closing the gap to prevent impact between the template and the substrate.
• the chuck is an air cavity chuck as discussed in U.S. Patent Nos. 6,982,783; 7,019,819 and 6,980,282, which are incorporated by reference herein in their entirety.
• step 104 film 204 is then cross-linked using UV light 206 as shown in Figure 2E. That is, film 204 is cured by cross-linking so as to polymerize film 204.
• step 105 template 205 is separated from substrate 202 to leave the polymerized film 207 on substrate 202 as shown in Figure 2F.
• template 205 has a multi-tiered pattern and/or a three-dimensional shape.
• template 205 is flexible or held in a roll-to-roll configuration. This template configuration can be realized after substrate 202 is patterned using a rigid template 205, and can be used to pattern large-area substrates, such as glass panels, as discussed in U.S. Patent No. 9,616,614, which is incorporated by reference herein in its entirety.
• a roll-to-roll PFIL approach can support both roll-to-roll and roll-to-plate (R2P) lithography.
• the former involves having both the template and the substrate held in roll-to-roll configurations, while the latter has one of the two held in a roll-to-roll configuration with the other being a rigid substrate, such as a wafer or a glass panel.
• the substrate is assumed to be held in a roll-to-roll configuration while the template is rigid.
• An example of roll-to-plate lithography to make replica templates, and final substrates, is provided in Figures 3 and 4A-4J.
• Figure 3 is a flowchart of method 300 for fabricating replica templates and final substrates using roll-to-plate lithography in accordance with an embodiment of the present invention.
• Figures 4A-4J depict the cross-sectional views of fabricating replica templates and final substrates using the steps described in Figure 3 in accordance with an embodiment of the present invention.
• a master template 401 is fabricated, such as using e-beam, photolithography, etc., as shown in Figure 4 A.
• master template 401 is slot-die coated with resist 402 as shown in Figure 4B.
• R2P roll-to-plate
• step 304 replica 405 is separated from master template 401 as shown in Figure 4D and the process is repeated (steps 302-304) to fabricate further replicas 405.
• step 305 oxide 406 is deposited on master template 401 as shown in Figure 4E.
• replica 405 is imprinted (R2P) on master template 401 with oxide 406 forming replica template 407 as shown in Figures 4E and 4F.
• Step 306 is repeated to fabricate a roll of replica templates 407 with a protective interleaf as shown in Figure 4F.
• a substrate 408 is slot-die coated with resist 409 as shown in Figure 4G.
• resist 409 is imprinted (R2P) using replica template 407 and cured as shown in Figure 4H.
• step 309 substrate 408 is separated from replica template 407 forming a patterned resist 410 on substrate 408 as shown in Figures 41 and 4J.
• the inverse model may also include parameters of the Digital Micromirror Device (DMD) array, for example, the spatial distribution and frequency of switching of individual mirrors, to control the spatial evaporation profile.
• DMD Digital Micromirror Device
• the inverse optimization scheme may be augmented to determine one of the following: a spatial evaporation profile, a spatial temperature profile, a spatial material removal profile, a spatial evaporation rate profile, and a spatial material removal rate profile.
• the inverse model should be wrapped around a forward model that takes into account thin film flow in the presence of evaporation and surface tension driven gradients for the drop spreading, merging and evaporation stages prior to imprinting.
• a framework for this optimization scheme is shown in Figure 5.
• Figure 5 is a diagram of the framework for the inverse optimization scheme for D-PFIL in accordance with an embodiment of the present invention.
• optimization solver 501 receives as inputs, the process parameters 502, topography 503, a desired pre-imprint residual layer thickness (RLT) profile 504, which uses the template pattern 505 as an input, and a predicted pre-imprint RLT profile 506.
• RLT residual layer thickness
• Optimization solver 501 outputs the inkjet material distribution 507, temperature distribution 508 and merging and spreading times 509.
• the initial volumes and locations 510, which receives as input, the inkjet material distribution 507, as well as temperature distribution 508 and merging and spreading times 509 are inputs to the thin film and evaporation model 511 which is inputted to predicted pre-imprint RLT profile 506.
• template pattern 505 is inputted to heuristics 512 which is inputted to initial volumes and locations 510.
• such an inverse optimization scheme is implemented by a computing system.
• any deviations in drop volume in the drop dispense stage can be overcome through image-processing based drop volume and velocity control.
• the inverse model for determining optimum drop volumes and locations can also leverage prior work from Sreenivasan et al. (U.S. Patent Application Publication No. 2015/0048050), which is incorporated by reference herein in its entirety.
• the boiling point of the solvent should be much lower than the rest of the resist components, such that there is minimal evaporation of the resist when evaporating the solvent.
• the boiling point of the XNTL-26SF resist from MicroResist Technologies is close to 180° C
• the boiling point of a common solvent, such as PGMEA is close to 140° C.
• This selective evaporation of the resist material can enable a desired film thickness profile, based on the need to obtain a uniform residual layer in the presence of non-uniform patterns, or address substantially systematic process parasitics, such as non-uniform shrinkage upon cross-linking and non-uniform etch rates in etch tools.
• the film thickness profile can also be used to obtain a desired functional performance, such as desired optical performance, mechanical stability, etc.
• a resist is partially evaporated in this manner, it is desirable that the relative evaporation rates of all the components of the resist are substantially the same to ensure the resist remains consistent in its properties. This can be done by keeping the boiling points of the components of the resist mixture within a range of 10° C with each other. For substrate planarization, however, the material design may involve additional considerations. Details of material design are discussed further below.
• spin-coating is the preferred material deposition technique.
• Spin-coating, followed by baking, is a commonly used technique for obtaining substantially uniform films on rigid substrates.
• the challenge with spin-coating is its inability to cater to non-uniform material requirement in the presence of template pattern variation, which can be overcome using SC-PFIL.
• An embodiment of the process flow for SC- PFIL is shown in Figure 6.
• Figure 6 is a flowchart of a method 600 of the spin-coated based P-FIL (SC-PFIL) process for fabricating a pattern in accordance with an embodiment of the present invention.
• Figure 6 will be discussed below in connection with Figures 7A-7F.
• Figures 7A-7F depict the cross-sectional views of fabricating a pattern using the steps described in Figure 6 in accordance with an embodiment of the present invention.
• step 601 an initial volume of diluted monomer 701 (diluted in a solvent) is deposited on substrate 702 forming a liquid resist formulation as shown in Figure 7A.
• step 602 solvent 703 is evaporated from the liquid resist formulation during spin- coating and baking forming a thin film of monomer 704 with little or no concentration of solvent as shown in Figures 7B and 7C.
• the liquid resist formulation consists of solvent with a substantially higher volatility than the rest of the components of the liquid resist formulation.
• the non-solvent component(s) of the liquid resist formulation have substantially similar volatility.
• the non-solvent component(s) of the liquid resist formulation except for a photoinitiator have a boiling point that differ by less than 30° C, 10° C or 5° C.
• the non-solvent component(s) include a thermoset polymer with a high glass transition temperature.
• the non-solvent component(s) include silicon or carbon.
• the thin film of monomer 704 is selectively evaporated, such as via infra-red (IR) lamps, forming a non-uniform and substantially continuous film 705 on substrate 702 as shown in Figure 7D.
• the selective evaporation is carried out by one or more of the following: a temperature controlled substrate chuck, an array of individually controllable infrared lamps, an array of individually controllable fans to control air flow, individually controllable micro heaters or coolers on substrate 702, and a spatial light modulator (may include one or more Digital Micromirror Device (DMD) arrays) to control one or more sources of electromagnetic radiation.
• DMD Digital Micromirror Device
• the solvent evaporation, selective resist evaporation and imprinting are carried out in stations that may be separated by barriers that prevent vapor transport between the stations.
• the selective evaporation is carried out in an environment with controlled vapor pressures.
• the liquid resist formulation, whose solvent has already been evaporated, which consists of the monomer film 704, is selectively evaporated to form film 705 with an intentionally non-uniform profile to match a desired thickness profile.
• step 604 a gap between a template 706 and substrate 702 is then closed (bringing template 706 down in contact with film 705) as shown in Figure 7D.
• template 706 is patterned with a lateral feature dimension of less than 500 nm. In one embodiment, template 706 is patterned with a lateral feature dimension of less than 100 nm.
• step 605 film 705 is then cross-linked using UV light 707 as shown in Figure 7E. That is, film 705 is cured by cross-linking so as to polymerize film 705.
• step 607 template 706 is separated from substrate 702 to leave the polymerized film 708 on substrate 702 as shown in Figure 7F.
• Figure 8 is a flowchart of a method 800 of an alternative embodiment of the spin-coated based P-FIL (SC-PFIL) process for fabricating a pattern in accordance with an embodiment of the present invention.
• Figure 8 will be discussed below in connection with Figures 9A-9F.
• Figures 9A-9F depict the cross-sectional views of fabricating a pattern using the steps described in Figure 8 in accordance with an embodiment of the present invention.
• step 801 an initial volume of diluted monomer 901 (diluted in a solvent) is deposited on substrate 902 forming a liquid resist formulation as shown in Figure 9A.
• step 802 solvent 903 is evaporated form the liquid resist formulation during spin- coating and baking forming a thin film of monomer 904 with little or no concentration of solvent as shown in Figures 9B and 9C.
• the liquid resist formulation consists of solvent with a substantially higher volatility than the rest of the components of the liquid resist formulation.
• the non-solvent component(s) of the liquid resist formulation have substantially similar volatility.
• the non-solvent component(s) of the liquid resist formulation except for a photoinitiator have a boiling point that differ by less than 30° C, 10° C or 5° C.
• the non-solvent component(s) include a thermoset polymer with a high glass transition temperature.
• the non-solvent component(s) include silicon or carbon.
• the thin film of monomer 904 is selectively evaporated, such as via infra-red (IR) lamps, forming a non-uniform and substantially continuous film 905 on substrate 902 as shown in Figure 9D.
• the selective evaporation is carried out by one or more of the following: a temperature controlled substrate chuck, an array of individually controllable infrared lamps, an array of individually controllable fans to control air flow, individually controllable micro heaters or coolers on substrate 902, and a spatial light modulator (may include one or more Digital Micromirror Device (DMD) arrays) to control one or more sources of electromagnetic radiation.
• DMD Digital Micromirror Device
• the selective evaporation and patterning are carried out in separate stations to avoid recondensation of evaporated materials onto substrate 902 prior to patterning.
• the selective evaporation is carried out in an environment with controlled vapor pressures.
• the liquid resist formulation, whose solvent has already been evaporated, which consists of the monomer film 904 is selectively evaporated to form film 905 with an intentionally non-uniform profile to match a desired thickness profile.
• step 804 a gap between a superstrate 906 and substrate 902 is then closed (bringing superstrate 906 down in contact with film 905) as shown in Figure 9D.
• superstrate 906 is used to cover the top of film 905 to obtain desired evolution dynamics of film 905.
• a superstrate thickness profile of superstrate 906 and geometry are optimized to obtain desired evolution dynamics for a given substrate topography.
• superstrate 906 is flexible or held in a roll-to-roll configuration.
• step 805 film 905 is then cross-linked using UV light 907 as shown in Figure 9E. That is, film 905 is cured by cross-linking so as to polymerize film 905.
• step 806 superstrate 906 is separated from substrate 902 to leave the polymerized film 908 on substrate 902 as shown in Figure 9F.
• the material can be substantially diluted using solvents, although the process can be executed using resist materials that are substantially free of solvents. Details of material design are discussed further below.
• the monomer resist material concentration can be minimized to reduce the waste of expensive material.
• the solvent 703, 903 evaporates leading to a thin film of monomer resist 704, 904 with little or no concentration of solvent. This evaporation during the spin-coating process can also be affected by the air flow in the vicinity.
• the coated film immediately after spin-coating is typically substantially uniform.
• the inverse model can be used to obtain optimum values for these parameters to obtain an initial uniform thick film (e.g., thickness of liquid resist formulation) that can then be driven to non-uniformity using controlled evaporation of the monomer in a physically separated station, similar to D-PFIL.
• the model can be augmented to account for evaporation of the resist which drives the final film thickness profile.
• the evaporating solvent is completely evaporated and exhausted before the substrate is carried to the patterning station, to prevent recondensation of the solvent on the substrate.
• Any residual solvent, either on the substrate or in the vapor phase, can create problems with voids, feature collapse or parasitic condensation on the patterned resist.
• a framework for inverse optimization in the SC-PFIL process is shown in Figure 10.
• Figure 10 is a diagram of the framework for the inverse optimization scheme for SC- PFIL in accordance with an embodiment of the present invention.
• optimization solver 1001 receives as inputs, the input parameters 1002, topography 1003, a desired pre-imprint residual layer thickness (RLT) profile 1004, which uses the template pattern 1005 as an input, and a predicted pre-imprint RLT profile 1006.
• RLT residual layer thickness
• Optimization solver 1001 outputs the initial volume of material 1007, temperature distribution 1008 and spin process parameters 1009.
• the film thickness after spin-coating 1010 which receives as input, the initial volume of material 1007, as well as temperature distribution 1008 and merging and spin process parameters 1009 are inputs to the thin film and evaporation model 1011 which is inputted to predicted pre-imprint RLT profile 1006.
• template pattern 1005 is inputted to heuristics 1012 which is inputted to film thickness after spin-coating 1010.
• such an inverse optimization scheme is implemented by a computing system.
• the control of resist evaporation can be done through judicious control of temperature through a distributed set of thermal actuators, IR lamps, or combinations thereof, in which each element in the set can be individually controlled.
• This individual control in IR lamps can be achieved through devices, such as Digital Micromirror Device (DMD) Arrays from Texas Instruments.
• DMD Digital Micromirror Device
• a state-of-the-art array has 4M pixels at a pitch of 7.6 micrometers at a diagonal width of 0.9 inches. This can enable spatial modulation of IR intensity to control the evaporation over at least an area of - 0.4 sq. inch. With the appropriate reduction optics, a larger area can be projected upon, but by compromising resolution.
• a resist formulation containing methyl methacrylate (MMA) as the liquid monomer and primary component would need a local heat source of ⁇ 1 W to absorb sufficient heat at normal temperature and pressure to evaporate -100 nm of liquid film from a local 50 x 50 ⁇ 2 area, assuming the starting film thickness is 500 nm.
• Such powers can be delivered using DMD arrays. These devices also have frame rates approaching 10 kHz, and thereby provide high frequency temporal control of the incident radiation and subsequent evaporation.
• the inverse model may also include parameters of the DMD array, for example, the spatial distribution and frequency of switching of individual mirrors, to control the spatial evaporation profile.
• multiple DMD arrays can be used across a larger area, if resolution cannot be compromised further.
• the individual pixels may have significant cross-talk, which can also be modeled and incorporated in the forward model for thin film evaporation.
• an underlying substrate with MEMS-based micro-heaters may also be used for enhanced temperature distribution, if a higher resolution is required than what can be afforded by the DMD arrays.
• This principle of selective heating can be generally applied to all embodiments of the present invention and is not specific to only SC-PFIL.
• the starting material can be substantially free from the solvent and may only consist of the monomer resist (e.g., XNTL26-SF resist from MicroResist Technologies).
• the approach presented herein uses controlled resist evaporation thereby removing a portion of the monomer mixture that is to be cross-linked.
• This selective evaporation of the resist material can enable a desired film thickness profile, based on the need to obtain a uniform residual layer in the presence of non-uniform patterns, or address substantially systematic process parasitics, such as non-uniform shrinkage upon cross-linking, and non-uniform etch rates in etch tools.
• the film thickness profile can also be used to obtain a desired functional performance, such as desired optical performance, mechanical stability, etc.
• the relative boiling points of all the components of the resist are substantially the same, for example, within 10° C of each other, to ensure the resist mixture remains consistent in its properties; and it is desirable to not include a solid material as a component of the resist mixture.
• the material design may involve additional considerations. Details of material design are discussed further below.
• slot die coating is the material deposition technique.
• Slot die coating is an inexpensive process that has been used traditionally for coating rectangular substrates, such as display glass.
• the challenge with slot die coating is that it is extremely difficult to obtain nanoscale wet films of the thickness desired for nanoscale imprint lithography process residual layer films that have a mean thickness of less than 50 nm.
• the material can be substantially diluted using solvents. Details of material design are discussed further below. Thus, the resist material concentration can be minimized to reduce the waste of expensive material.
• the coated film immediately after slot die coating is substantially uniform in at least one dimension.
• process parameters that leads to a desired mean film thickness (e.g., desired thickness of liquid resist formulation) after slot die coating.
• these control parameters include speed of the substrate relative to the die and coating gap.
• the inverse model can be used to obtain optimum values for these parameters to obtain an initial uniform thick film that can then be driven to non-uniformity using controlled evaporation of the remaining resist in a separate station.
• the relation between these parameters and the film thickness can be quite complex to model. This is because the relation depends on the "region of operation" of the slot die coater. Typically, it would be preferred to have the coater operate in the so-called "region III," where the wet film thickness decreases with increasing speed, leading to favorable throughput. However, there might be situations where the material rheology may constrain the operation to only outside this region.
• the inverse model may be augmented with heuristics to determine an optimum set of values.
• FIG 11 is a diagram of the framework for the inverse optimization scheme for SD- PFIL in accordance with an embodiment of the present invention.
• optimization solver 1101 receives as inputs, the input parameters 1102, topography 1103, a desired pre-imprint residual layer thickness (RLT) profile 1104, which uses the template pattern 1105 as an input, and a predicted pre-imprint RLT profile 1106.
• RLT residual layer thickness
• Optimization solver 1101 outputs the region of operation 1107, temperature distribution 1108 and slot die process parameters 1109.
• the film thickness after slot die coating 1110, which receives as input, the region of operation 1107, as well as temperature distribution 1108 and slot die process parameters 1109 are inputs to the thin film and evaporation model 1111 which is inputted to predicted pre-imprint RLT profile 1106.
• template pattern 1105 is inputted to heuristics 1112 which is inputted to film thickness after slot die coating 1110.
• such an inverse optimization scheme is implemented by a computing system.
• the control of resist evaporation can be done through judicious control of temperature through a distributed set of thermal actuators, IR lamps, or combinations thereof, in which each element in the set can be individually controlled.
• This individual control in IR lamps can be achieved through devices, such as Digital Micromirror Device (DMD) Arrays from Texas Instruments.
• DMD Digital Micromirror Device
• a state-of-the-art array has 4M pixels at a pitch of 7.6 micrometers at a diagonal width of 0.9 inches. This can enable spatial modulation of IR intensity to control the evaporation over at least an area of - 0.4 sq. inch. With the appropriate reduction optics, a larger area can be projected upon, but by compromising resolution.
• a resist formulation containing methyl methacrylate (MMA) as the liquid monomer and primary component would need a local heat source of ⁇ 1 W to absorb sufficient heat at normal temperature and pressure to evaporate -100 nm of liquid film from a local 50 x 50 ⁇ 2 area, assuming the starting film thickness is 500 nm.
• Such powers can be delivered using DMD arrays. These devices also have frame rates approaching 10 kHz, and thereby provide high frequency temporal control of the incident radiation and subsequent evaporation.
• the inverse model may also include parameters of the DMD array, for example, the spatial distribution and frequency of switching of individual mirrors, to control the spatial evaporation profile.
• the inverse optimization framework compensates for sequentially selective evaporation from one or more DMD arrays, or temperature profiles resulting from heat capacity of the substrate or material, or deterministic changes in film thickness profile due to known changes in ambient conditions or when handling or transporting the substrate.
• the individual pixels may also have significant cross-talk, which can also be modeled and incorporated in the forward model for thin film evaporation.
• an underlying substrate with MEMS-based micro-heaters may also be used for enhanced temperature distribution, if a higher resolution is required than what can be afforded by the DMD arrays.
• the starting material can be substantially free from the solvent and may only consist of the resist (e.g., XNIL26-SF resist from MicroResist Technologies). The approach presented herein uses controlled resist evaporation thereby removing a portion of the monomer mixture that is to be cross-linked.
• This selective evaporation of the resist material can enable a desired film thickness profile, based on the need to obtain a uniform residual layer in the presence of non-uniform patterns, or address substantially systematic process parasitics, such as non-uniform shrinkage upon cross-linking, and nonuniform etch rates in etch tools.
• the film thickness profile can also be used to obtain a desired functional performance, such as desired optical performance, mechanical stability, etc.
• a resist mixture is partially evaporated in this manner, it is desirable that the relative boiling points of all the components of the resist are substantially the same, for example, within 10° C of each other, to ensure the resist mixture remains consistent in its properties; and it is desirable to not include a solid monomer material as a component of the monomer mixture.
• the material design may involve additional considerations. Details of material design are discussed further below.
• resist materials diluted by solvent have lower viscosity than those that are undiluted, as seen in the XNIL26 family of resists from MicroResist Technologies.
• the diluted materials are also more volatile.
• this equation can be used to derive desirable material properties to achieve a desirable process time scale.
• uniform width lines are routinely formed by merging multiple drops.
• a stability criterion has been established to determine when such lines are stable and when they lead to defects, as shown in Figure 12 discussed below.
• Such heuristics can be used to tweak drop pitch to enable stable film formation, assuming there is no evaporation in the spreading and merging stages.
• the process can be repeated after solvent evaporation, to get defined stable lines over the continuous film that are parallel to the directionality of the pattern.
• Figure 12 illustrates a stable bead formation in accordance with an embodiment of the present invention.
• the x-axis of Figure 12 is a function of drop spacing and contact angle, while the y-axis is a function of printhead velocity and fluid material properties.
• FIG. 13 illustrates variation of film thicknesses with spin time and spin speed in accordance with an embodiment of the present invention.
• the spin-coating process has three stages: deposition and spin-up, spin-off and film drying. The first two stages are not amenable for process control but the third stage, where a continuous film has been formed and the film thickness is continuously decreasing because of evaporation from the free surface, does offer some potential for control.
• the second is a journal article (Romero et al., "Response of Slot Coating Flows to Periodic Disturbances," Chemical Engineering Science, Volume 63, Issue 8, April 2008, Pages 2161-2173, ISSN 0009-2509, http://dx.doi.Org/10.1016/j .ces.2008.01.012) where the authors have conducted a perturbation analysis for the final wet film thickness on coating gap and upstream pressure variations. They report that the wet film thickness variation has a phase shift from the coating gap variation.
• the film thickness variation amplitudes are usually damped, but at high perturbation frequencies ( ⁇ 1 KHz), there can be high-amplitude oscillations in the wet film thickness.
• the slot die coating process too, can benefit from a prior experimental determination of the process window that can be used to augment the model in the form of heuristics, which can also drive tweaking of the film thickness profile to accommodate directional patterns on the template.
• in-line metrology is used to provide feedback for optimization using a machine in a loop control.
• the inline metrology uses one or more of the following: visual feedback, optical feedback, electromagnetic feedback and thermal feedback.
• the in-line metrology is performed using an array of microelectromechanical systems (MEMS)-based Atomic Force Microscopes (AFMs) with individual control of a location of each AFM tip.
• the array of AFM tips has individually addressable locations.
• an optimum positioning of the array of AFM tips on a wafer with a pattern, a topography variation or a variation in a thickness of a film(s) is determined.
• Each AFM tip in the array of AFM tips provides data that is aggregated to infer a pattern, a topography variation or a film thickness variation.
• the wafer may be scanned to measure the pattern, topography variation or film thickness variation.
• the measured pattern, topography variation or film thickness variation is aggregated and analyzed to infer the pattern, topography variation or film thickness variation.
• each AFM tip of the array of AFM tips is on a different device on the same wafer. In one embodiment, each of these different devices is the same.
• the process metrics that ensure reliable coating include mean and standard deviation of coating thickness over large areas, and avoidance of defect mechanisms.
• Current off-line metrology approaches are too slow to be incorporated in real-time process control as nanoscale measurements have to be made over centimeter scales resulting in a high-dimensional multi-scale metrology problem.
• New sensing architectures that are designed in conjunction with process models can be used to significantly enhance the speed of sensing and detection of process parasitics and defects.
• Optical techniques such as film thickness reflectometry, ellipsometry, spectrometry, etc.
• the sensing can happen very close to the nozzle and before the drop sits on the substrate.
• a camera can be collocated with the die to capture the coating bead formation.
• the coating gap is much smaller ( ⁇ 10 X) than the distance between the nozzle and the substrate in a typical inkjetting environment.
• Planarization of feature transitions presents a challenge. This is because, such sharp transitions entail a change in fluid and solid mechanical behavior, which is difficult to characterize in robust models, and which require in-situ validation. Moreover, the use of high throughput optical film thickness metrology, which works well for bulk films without sharp thickness changes within the optical spot size, is not compatible with measuring film thickness changes in the immediate vicinity of such steps. This entails the use of relatively slow tip-based surface profiling techniques, such as atomic force microscopy (AFM).
• AFM atomic force microscopy
• the use of an array of AFMs with the ability to automatically reconfigure individual AFM tip locations and scan regions is presented.
• One embodiment for creating these automatically reconfigurable arrays is to use a Si-micro chip based AFM MEMS device. The use of an AFM for offline characterization in production environments for semiconductor fabrication is provided by Bruker Sciences.
• the present invention extends the idea of the AFM array for planarization with the ability to reconfigure the location of each AFM tip in the array based on the pattern layout across a full field (typically 26 mm x 33 mm), typically obtained from a GDS file. More specifically, knowing the GDS file of the prior pattern(s), the prior pattern's etch depth and any subsequent film deposition on that pattern, a knowledge of the specific regions where planarization would be a challenge, could be inferred a priori through empirical or computational models, or a combination of the two. The AFM array could then be reconfigured to specifically measure in those challenging regions. This way, the AFM array can rapidly measure the film thickness profile in the vicinity of critical step transitions.
• the AFM array can be devised such that it is reconfigurable across an entire wafer, rather than a single field. This is because most fields in a silicon wafer undergoing routine lithography processing, are substantially similar.
• the AFM array can then be arranged such that a single AFM tip has the range to address a single field. Different AFM tips can then measure different features on the substantially similar field, with each feature being on a different field. Aggregating these measurements can then provide a picture of the surface of a field. If a different field layout is then processed, the same AFM array can be reconfigured to enable metrology of the new layout. This embodiment is applicable to U.S. Patent No. 9,415,418, which is incorporated by reference herein in its entirety.
• machine-in-the-loop forward models, validated numerical simulations and stochastic optimal control techniques based on genetic algorithms can be used to develop sophisticated control algorithms for the P-FIL process.
• a hierarchical approach can be used to gradually increase the number of control variables to be included in the optimized control algorithm.
• a sensing and actuator framework of low complexity can be first used to develop GA based process control schemes.
• the complexity of sensing (e.g., the spatial and spectral resolution of optical sensing) and actuator framework e.g., number and resolution of control variables
• the GA based process control schemes can be established.
• the formulation design consists of imprint resist comprising of considerable amount of components having similar volatility, dissolved in a solvent that is considerably more volatile than the rest of the components.
• the solvent role is to dilute the required quantity of imprint resist to higher volume thus allowing for better way of spreading small amounts of imprint resist over the substrate in IJ-PFIL, and managing the final dry film thickness starting from a much thicker initial wet film in SC-PFIL or SD-PFIL.
• the remaining, solvent- depleted liquid is then selectively evaporated leaving a substrate containing areas with controlled imprint resist distribution.
• the final ratio of the remaining components may not be the same as the initial ratio of the same components in the resist formulation.
• the components ratios may not be the same on various spots across the substrate after selective evaporation.
• appropriately formulated systems may perform satisfactorily under a range of compositions. Formulation design may take into account the expected amount of material that needs to be evaporated, such that the optimum range of component ratios is not disrupted. Further, certain components, such as the photoinitiator or crosslinkers, will almost always be less volatile, as seen in the table below.
• an imprint resist formulation may contain a mixture of some, or all the following components: an initiator; polymerizable monomers with one active group; polymerizable monomers with more than one active group referred to in the art as crosslinkers; and surfactants. This list is not exhaustive as other components may be present according to desired performance and applications.
• Downstream process considerations may preclude the use of UV-crosslinkable polymers, particularly in the application of planarization, which does not require a patterning resist material.
• most methacrylate polymers have a Tg of ⁇ 150° C. Any downstream thermal processing beyond 150° C may damage the material.
• thermoset polymers such as polyimide, that have a much higher Tg of -350° C.
• planarization using this polymer can alleviate concerns with thermal processing. But these materials are also highly viscous, which makes them difficult to coat or inkjet with substantially lower volumes.
• the polyimide material can be diluted in a solvent, which can reduce their viscosity and make the material amenable to coating and/or inkjetting.
• the solvent will need to be completely evaporated to minimize condensation on the substrate.
• a solvent with substantially higher volatility e.g., MIBK, ethyl acetate
• the polymer can then be baked or cured on the substrate.
• the liquid formulation (or "liquid resist formulation") used for this purpose is also called a "resist," having a solvent component and one or more non-solvent components.
• one embodiment of the invention can involve the use of partially evacuated chambers to reduce the boiling points of the resist components.
• Another embodiment can involve the use of a chuck or mount for the substrate which is substantially maintained at a constant temperature which is below the boiling points of the resist components.
• This chuck or mount can be metallic to enable rapid dissipation of any heat that is transferred to it from the substrate and thereby allow the substrate to substantially maintain its temperature, even in the presence of dynamic heat loads during the selective evaporation process.
• Another embodiment of the invention involves the use of one or more material films between the substrate and the resist material.
• films could be chosen such that their combination would lead to near zero transmission of electro-magnetic radiation in the infrared spectrum to the substrate, thereby preventing the substrate from heating, but can themselves absorb the heat without suffering any damage.
• An exemplar film material is CVD Carbon or Spin-on-Carbon. These materials may additionally also be used as sacrificial films when transferring the resist pattern onto the substrate.
• planarization includes the use of silicon or carbon containing materials that facilitate good mechanical properties, such as elastic modulus, elongation to fail and failure strength.
• the embodiment of free-surface planarization can use a low- viscosity silicon-containing resist material as discussed in Lin, M.W., "Simulation and design of planarizing materials and interfacial adhesion studies for step and flash imprint lithography", Ph.D. Dissertation, The University of Texas at Austin, 2008. This material can enable a high initial degree of planarization which can be further improved through selective evaporation of the resist material.
• nanophotonic devices include wire-grid polarizers and metamaterials, such as flat lenses. These devices typically involve fabrication of nanostructures made of metals (Al, Au, etc.) or dielectrics (Si0 2 , S13N4, Ti0 2 , etc.).
• wire-grid polarizers for display applications require nanoscale patterns made of aluminum.
• Such polarizers can be fabricated using one of two approaches. In the first approach, a polymer resist line/space pattern can be obtained using the PFIL process. This can be followed by Glancing Angle Deposition of Al on the line/space pattern such that the metal remains discontinuous.
• the second approach involves deposition of an Al film using vacuum-based processes, such as e- beam evaporation, followed by the imprinting of a polymer resist line/space pattern using PFIL. This is then followed by a descum etch of the residual layer using 0 2 plasma and etching of the underlying Al film using Cl-based plasma chemistry.
• Metamaterials such as flat lenses, have been discussed in the context of enabling interaction of visible light with nanoscale patterns to achieve phenomena that are otherwise impossible in the absence of nanoscale patterns.
• Flat lenses for example, enable the 'refraction' of light from flat surfaces, that would otherwise have required some curvature. These lenses are enabled by fabricating specific nanoscale patterns on the flat surface that interact with incoming light to result in refraction. These nanoscale patterns are made of high-index dielectric materials, most notable, Ti0 2 .
• Such flat lenses can be made using PFIL in combination with other nanoscale fabrication techniques.
• One approach for the fabrication of flat lenses involves imprinting the negative of the desired Ti0 2 pattern with a polymerized resist using PFIL.
• This continuous film of the Ti0 2 precursor can be etched using reactive ion etching in a 0 2 /Ar/F based plasma.
• the features on this resist pattern can be modified at the top to make them tapered.
• Such structures can be fabricated using multi- tiered templates fabricated for imprint lithography. This would allow the sol-gel Ti0 2 precursor to substantially fill the resist mold without forming a substantially continuous film at the top, provided the volume of the precursor is tightly controlled in the slot-die coating process.
• Another approach can involve conformal Chemical Vapor Deposition or Atomic Layer Deposition of Ti0 2 to fill the resist mold, followed by an etch-back of the continuous Ti0 2 film at the top of the mold.
• the resist pattern can be removed using 0 2 plasma.
• Figure 14 is a flowchart of a method 1400 for fabricating nanoscale patterns made of high-index dielectric materials, such as Ti0 2 , on a flat surface in accordance with an embodiment of the present invention.
• Figures 15A-15D depict the cross-sectional views of fabricating nanoscale patterns on a flat surface using the steps described in Figure 14 in accordance with an embodiment of the present invention.
• resist 1501 is patterned on a substrate 1502 as shown in Figure 15 A.
• step 1402 a short descum ash is performed to expose substrate 1502 as shown in Figure 15B.
• substrate 1502 is slot-die coated with Ti0 2 1503 or, alternatively, Ti0 2 1503 is deposited on substrate 1502 via chemical vapor deposition as shown in Figure 15C.
• step 1404 a blanket etch of Ti0 2 1503 is performed as shown in Figure 15D.

Landscapes

• General Physics & Mathematics (AREA)
• Physics & Mathematics (AREA)
• Engineering & Computer Science (AREA)
• Chemical & Material Sciences (AREA)
• Nanotechnology (AREA)
• General Health & Medical Sciences (AREA)
• Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
• Radiology & Medical Imaging (AREA)
• Health & Medical Sciences (AREA)
• Condensed Matter Physics & Semiconductors (AREA)
• Manufacturing & Machinery (AREA)
• Crystallography & Structural Chemistry (AREA)
• Shaping Of Tube Ends By Bending Or Straightening (AREA)
• Exposure Of Semiconductors, Excluding Electron Or Ion Beam Exposure (AREA)
• Application Of Or Painting With Fluid Materials (AREA)

Priority Applications (2)

Applications Claiming Priority (2)

Publications (1)

Family

ID=61073232

Family Applications (2)

Family Applications After (1)

Country Status (4)

Cited By (4)

Families Citing this family (6)

Citations (11)

Family Cites Families (28)

• 2017
  • 2017-08-03 WO PCT/US2017/045373 patent/WO2018027073A1/en not_active Ceased
  • 2017-08-03 WO PCT/US2017/045368 patent/WO2018027069A1/en not_active Ceased
  • 2017-08-03 EP EP17837715.6A patent/EP3493921A4/en active Pending
  • 2017-08-03 JP JP2019506101A patent/JP7041121B2/ja active Active
  • 2017-08-03 US US16/322,882 patent/US11762284B2/en active Active
  • 2017-08-03 US US16/322,842 patent/US11669009B2/en active Active

Patent Citations (11)

Also Published As

Similar Documents

Legal Events