Imprinting method, imprinting lithography system, and article manufacturing method

Abstract

The present invention relates to imprinting methods, imprint lithography systems, and article manufacturing methods. The present disclosure provides methods and systems for imprinting a substrate supported on a movable stage, including contacting a liquid resist on the substrate with a template, and aligning the substrate with the template using a feedback- feedforward control process using a first set of control parameters. Calibration data is obtained, and a viscosity of at least a portion of the resist is increased. After increasing the viscosity of the resist, the substrate is aligned with the template using the feedback- feedforward control process using a second set of control parameters, the second set of control parameters is determined based on the calibration data, and the resist under the template is solidified.

Description

Technical Field

[0001] This disclosure relates to alignment control in nanoimprint lithography, and more specifically to compensation for vibrations introduced during the alignment process. Background Technology

[0002] In nanoimprint lithography, field-to-field alignment techniques have been used to achieve nanoscale overlay accuracy. In some examples, initial alignment errors between the imprint template and corresponding areas on the substrate can be corrected by moving the template relative to the substrate (e.g., a wafer). However, rapid and consistent alignment in nanoimprint lithography remains a challenge. More specifically, thin-film friction and variations in initial conditions are two major difficulties. This includes the difficulty of current schemes in handling varying RLTs (residual thickness of the curable liquid layer between the template and the substrate), positions, and transition processes. In particular, the distribution of RLTs, alignment movements, and vibrations generated by the lithography environment itself lead to unacceptable oscillation levels. Furthermore, due to the introduction of these varying vibration instances, it is impossible to zero out the oscillation levels through natural friction. These issues have consistently impacted yield and efficiency in large-scale production, thus requiring solutions to address them. Summary of the Invention

[0003] According to this disclosure, a method and system for imprinting a substrate supported on a movable stage are provided. The method and system include: contacting a liquid photoresist on the substrate with a template; aligning the substrate with the template using a feedback-feedforward control process employing a first set of control parameters; obtaining calibration data; and increasing the viscosity of at least a portion of the photoresist; after increasing the viscosity of the photoresist, aligning the substrate with the template using the feedback-feedforward control process employing a second set of control parameters; determining the second set of control parameters based on the calibration data; and curing the photoresist beneath the template.

[0004] According to another embodiment, the calibration data is generated while aligning the substrate and the template using the first set of control parameters and the feedback-feedforward control processing. In another embodiment, the first set of control parameters includes a frequency component, the second set of control parameters represents an updated frequency component generated based on the calibration data, and the substrate and the template are aligned based on the updated frequency component in the second set of control parameters.

[0005] In another embodiment, the calibration data is generated by: receiving a set of alignment controls that change over time; converting the set of alignment controls into a set of alignment state values ​​that change with a first state variable; receiving a set of position estimates that change over time; converting the set of position estimates into a set of position state values ​​that change with the first state variable; and generating one or more covariance values.

[0006] In one embodiment, the set of alignment controls includes: one or more adaptive functions used in the feedback-feedforward control process; one or more output control signals used to control the position of the stage; and an energy controller signal that controls one or more parameters for applying energy to increase the viscosity of the resist. In one embodiment, the set of position estimates includes: a first position signal representing the position of a mark on the template relative to a mark on the substrate, and a second position signal representing the position of the stage.

[0007] In another embodiment, generating the calibration data further includes obtaining a second set of alignment controls and a second set of position estimates, wherein the second set of alignment controls and the second set of position estimates are based on a previously performed imprinting process. In yet another embodiment, the generated covariance values ​​are used to generate the second set of control parameters by modifying one or more control functions applied during the feedback-feedforward control process.

[0008] According to another aspect of this disclosure, an imprint lithography system is provided for controlling the alignment of an imprint lithography template relative to a substrate. The system includes: a stage configured to hold the substrate and movable such that the position of the stage can be modified; a depositor configured to deposit a resist on the substrate; an energy controller configured to apply energy to the resist to increase its viscosity; and a sensor configured to sense the position of the substrate relative to the imprint lithography template. At least one controller, communicating with the stage and the sensor, is configured to, based on contact between the substrate having liquid imprint lithography resist and the template, perform the following operations: aligning the substrate with the template using a feedback-feedforward control process with a first set of control parameters; obtaining calibration data; increasing the viscosity of at least a portion of the resist; after increasing the viscosity of the resist, aligning the substrate with the template using the feedback-feedforward control process with a second set of control parameters, wherein the second set of control parameters is determined based on the calibration data; and curing the resist beneath the template.

[0009] According to another aspect of this disclosure, a method for manufacturing an article is provided, including the use of the above-described imprinting method, and the method further includes: dispensing an imprinting resist onto the substrate, the imprinting resist being a liquid; contacting the imprinting resist with an object, the object having a pattern in contact with the imprinting resist; and processing the substrate with the imprinting resist dispensed to manufacture an article. In another embodiment, processing the substrate further includes: applying energy to the substrate to cure the imprinting resist and forming a pattern on the substrate corresponding to the pattern on the object, wherein the imprinting method is repeatedly performed while the object is in contact with the imprinting resist, such that the object is aligned with the substrate before the resist is cured by the applied energy.

[0010] Details of one or more implementations of the subject matter described in this specification are set forth in the accompanying drawings and the description below. Other potential features, aspects, and advantages of the subject matter will become apparent from the specification, drawings, and claims. Attached Figure Description

[0011] Figure 1 A side view of the nanoimprint lithography system is depicted.

[0012] Figure 2 Depicting Figure 1 Side view of the substrate.

[0013] Figure 3 A side view of a nanoimprint lithography template in contact with a liquid imprint resist on a substrate is depicted, showing the initial alignment error X0 between example pairs of alignment marks on the template and the substrate, respectively.

[0014] Figure 4 A block diagram illustrating feedforward and feedback controls for aligning marks on the template and substrate is depicted.

[0015] Figure 5 The controlled viscosity alignment operation is shown.

[0016] Figure 6 This is a flowchart detailing the alignment control algorithm. Detailed Implementation

[0017] Figure 1An imprint lithography system 100 is shown that forms an embossed pattern on a substrate 102. The substrate 102 can be coupled to a substrate chuck 104. In some examples, the substrate chuck 104 includes a vacuum chuck, a pin chuck, a slot chuck, an electromagnetic chuck, or other suitable chuck. An exemplary chuck is described in U.S. Patent No. 6,873,087, which is incorporated herein by reference. The substrate 102 and the substrate chuck 104 can also be supported by a stage 106. The stage 106 provides movement about the X, Y, and Z axes, and rotation about the Z axis (e.g., θ). In this respect, the stage 106 can refer to an XYθ stage. The stage 106, the substrate 102, and the substrate chuck 104 can also be positioned on a base (not shown).

[0018] The imprint lithography system 100 includes an imprint lithography template 108 spaced apart from a substrate 102. In some examples, the template 108 includes a mesa 110 (mold 110) extending from the template 108 toward the substrate 102. In some examples, the mold 110 includes a patterned surface 112. The template 108 and / or the mold 110 may be formed of materials including, but not limited to, fused silica, quartz, silicon, organic polymers, siloxane polymers, borosilicate glass, fluorocarbon polymers, metals, hardened sapphire, or other suitable materials. In the illustrated example, the patterned surface 112 includes a plurality of features defined by spaced-apart grooves 124 and protrusions 126. The pattern formed as described above is merely an example, and any type of pattern can be presented on the patterned surface 112. Therefore, the patterned surface 112 can define any pattern that forms the basis for the pattern to be formed on the substrate 102 via the imprinting process.

[0019] Template 108 may be coupled to template chuck 128. In some examples, template chuck 128 includes a vacuum chuck, needle chuck, slot chuck, electromagnetic chuck, or any suitable chuck. Exemplary chucks are described in U.S. Patent No. 6,873,087. In some embodiments, template chuck 128 may be of the same type as substrate chuck 104. In other embodiments, template chuck 128 and substrate chuck may be different types of chucks. Furthermore, template chuck 128 may be coupled to impression head 130 such that template chuck 128, impression head 130, or both are configured to facilitate movement of template 108. Movement of template 108 includes movement within the template plane (in-plane movement) and movement outside the template plane (out-of-plane movement) relative to the substrate. In-plane movement includes the template 108 moving within the template plane (e.g., in a position such as...). Figure 1Translation of the template 108 within the XY plane (as depicted in the diagram) and rotation of the template within the template plane (e.g., within the XY plane and about the Z-axis). Translation or rotation of the template 108 relative to the substrate 102 can also be achieved by translation or rotation of the substrate. In-plane movement of the template 108 also includes increasing or decreasing the compressive force on opposite sides of the template (e.g., using an amplification actuator) to increase or decrease the size of the template within the XY plane of the template. Out-of-plane movement of the template 108 includes translation of the template along the Z-axis (e.g., increasing or decreasing the force applied to the substrate via the template by increasing or decreasing the distance between the template and the substrate) and rotation of the template about an axis within the XY plane of the template. Rotation of the template 108 about an axis within the XY plane of the template changes the angle between the XY plane of the template 108 and the XY plane of the substrate 102, referred herein as the “tilt” of the template relative to the substrate, or a change in the “tilt degree” or “tilt angle” of the template relative to the substrate. U.S. Patent No. 8,387,482 discloses the movement of a template via an imprint head in an imprint lithography system, which is incorporated herein by reference.

[0020] The imprint lithography system 100 may further include a fluid distribution system 132. The fluid distribution system 132 can be used to deposit a polymerizable material 134 on a substrate 102. The polymerizable material 134 can be arranged on the substrate 102 using techniques such as drop casting, spin coating, dip coating, chemical vapor deposition (CVD), physical vapor deposition (PVD), thin film deposition, thick film deposition, or other suitable methods. In some examples, the polymerizable material 134 is arranged on the substrate 102 before or after defining a desired volume between the mold 110 and the substrate 102. The polymerizable material 134 may include monomers as described in U.S. Patent No. 7,157,036 and U.S. Patent Application Publication No. 2005 / 0187339, both of which are incorporated herein by reference. In some examples, the polymerizable material 134 is arranged on the substrate 102 as a plurality of droplets 136.

[0021] Reference Figure 1 and Figure 2 The imprint lithography system 100 may also include an energy source 138 coupled to the direct energy 140 along path 142. In some examples, the imprint head 130 and stage 106 are configured to position the template 108 and substrate 102 in a manner overlapping with path 142. The imprint lithography system 100 may be regulated by a controller 144 communicating with the stage 106, imprint head 130, fluid distribution system 132, energy source 138, or any combination thereof, and may be operated according to a computer-readable program stored in memory 146.

[0022] In some examples, the impression head 130, stage 106, or both alter the distance between the mold 110 and the substrate 102 to define the desired volume therebetween filled with the polymerizable material 134. For example, the impression head 130 may apply force to the template 108, causing the mold 110 to contact the polymerizable material 134. After the desired volume is filled with the polymerizable material 134, an energy source 138 generates energy 140 (such as broadband ultraviolet radiation) to polymerize the polymerizable material 134 and conform it to the shape of the surface 148 and patterned surface 112 of the substrate 102, defining a polymerized patterned layer 150 on the substrate 102. In some examples, the patterned layer 150 includes a residual layer 152 and multiple features shown as protrusions 154 and grooves 156, wherein the protrusions 154 have a thickness t1 and the residual layer 152 has a thickness t2.

[0023] The aforementioned systems and processes can also be implemented in imprint lithography processes and systems mentioned in U.S. Patent No. 6,932,934, U.S. Patent Application Publication No. 2004 / 0124566, U.S. Patent Application Publication No. 2004 / 0188381, and U.S. Patent Application Publication No. 2004 / 0211754, all of which are incorporated herein by reference.

[0024] Imprint lithography substrates and stencils may include corresponding pairs of alignment marks that allow for real-time alignment of the stencil and substrate. After the patterned stencil is positioned on the substrate (e.g., stacked on the substrate), the alignment of the stencil alignment marks relative to the substrate alignment marks is determined. Alignment schemes may include performing “through-the-table” (TTM) measurements on alignment errors associated with the corresponding alignment mark pairs, and then compensating for these errors to achieve precise alignment of the stencil with the desired imprint location on the substrate, as disclosed in U.S. Patents 6,916,585; 7,170,589; 7,298,456; and 7,420,654, all of which are incorporated herein by reference. Alignment errors may be caused by the relative positioning of the substrate and stencil, deformation of the substrate or stencil, or a combination thereof. Alignment errors may also be caused by the introduction of vibrations from one or more actions of the imprint lithography process and the machine performing the imprint lithography process.

[0025] Figure 3A side view of an imprint lithography template 108 in contact with a liquid imprint resist 134 on a substrate 102 is shown, illustrating a first or initial alignment error X0 between an example pair of alignment marks 302 and 304 on the template 108 and the substrate 102, respectively. The alignment error X0 can be measured by an imaging device such as sensor 158. In some examples, sensor 158 includes a TTM aligner configured to detect diffracted light from alignment marks 302 and 304, where the diffracted light can pass through the liquid imprint resist 134. The initial alignment error X0 may exceed a tolerable alignment error, for example, less than 10 nanometers, with a repeatability error of 1 nanometer or less. While sensor 158 is described as an imaging device, this is merely exemplary, and the imaging device can include any device capable of detecting, capturing, and transmitting diffracted light in real time.

[0026] Alignment error X0 may be primarily caused by placement error, rotation error, and / or the compliance and hysteresis of stage 106 (e.g., XYθ stage), and may include errors in the X and Y axes as well as rotation (θ) about the Z axis. For example, placement error generally refers to the XY positioning error between the template and the substrate (i.e., translation along the X, Y, or both axes, where the X and Y axes lie in the plane of the imprinting surface of the template or substrate (or are parallel to the imprinting surface of the template or substrate), such as... Figure 1 The rotation (θ) error generally refers to the relative orientation error about the Z-axis (i.e., rotation about the Z-axis, where the Z-axis is orthogonal to the XY plane, such as...). Figure 1 (As depicted).

[0027] The placement error caused by the offset of the template alignment mark 302 and the corresponding substrate alignment mark 304 in the XY plane can be compensated by the relative movement of the template and the substrate (e.g., by controlled movement of the substrate, template, or both in the XY plane). The rotation error can be compensated by changing the relative angle between the template and the substrate in the XY plane (e.g., by rotation of the substrate, template, or both).

[0028] This disclosure proposes a control mechanism for controlling the above references. Figures 1 to 3The described imprinting system operates to reduce the amount of time required for alignment marks 302 and 304 to align properly, thereby ensuring that the pattern defined on the platform 110 is successfully imprinted onto the substrate 102. In other words, a control algorithm will be described below to keep the error value, typically indicated by X0, below a predetermined error threshold for a predetermined time period. Preferably, this will result in the relative distance between the marks on the template and the marks on the substrate being less than a predetermined distance value. However, due to the various physical properties of both the substrate and the polymer used to imprint the pattern thereon, it is difficult to achieve an acceptable level of error value within an acceptable time period.

[0029] One challenge associated with proper alignment between the substrate and the template involves vibrations introduced during the imprinting process that affect proper alignment. Vibrations introduced during imprinting have various sources, but isolating, identifying, and controlling the imprinting process to correct these vibrations is not always straightforward. Exemplary vibration sources may include, but are not limited to: resist application to the substrate, specific alignment vibrations caused by the alignment process, and vibrations caused by… Figure 1 Vibrations caused by any number of components of the imprint lithography machine shown.

[0030] An exemplary correction mechanism has been developed to minimize specific vibrations that may affect alignment. This correction mechanism, known as Controlled Viscosity Alignment (CVA), reduces vibration by increasing friction between the template and substrate by applying a predetermined amount of direct energy to the polymerizable material deposited onto the substrate via a dispensing mechanism. Both the intensity and optical wavelength of the energy directed at the polymerizable material vary spatially and temporally. CVA treatment has been found to minimize the effects of vibration. However, different conditions at different points in the photolithography process can lead to changes in the level and type of vibration. More specifically, vibration variations can be any one or more variations in amplitude, frequency, phase, and the time typically required for vibration to gradually disappear (fade-out envelope) due to natural friction. For example, CVA may be able to eliminate vibrations caused by the dispensing of the polymerizable material, environmental factors, and alignment motion, but the result of CVA treatment can be residual vibrations, in some cases, with frequencies different from the original vibrations corrected by CVA. In another example, CVA applied as described above to correct vibrations results in frictional uncertainty between the template and substrate, which can lead to frictional stalls in the alignment process.

[0031] In some cases, vibration variations after CVA may be related to different residual layer thicknesses (RLT), the current imprinting location (e.g., center, edge, partial area), and timing during alignment movements. These variations directly affect the CVA process, which is designed to minimize vibration. The presence of feedback control loops associated with the stage encoder and moiré sensor enables some vibration cancellation actions. However, other vibration sources are difficult to identify, thus posing challenges when attempting to control other components of the imprint lithography process to perform actions that can eliminate additional vibrations. Therefore, being able to tune the CVA process to better account for and correct for the variable vibrations introduced during imprint lithography would be particularly advantageous.

[0032] Therefore, a vibration correction algorithm is provided for use in a feedforward-feedback control loop for controlling an imprint lithography system. This algorithm advantageously considers multiple output signals and multiple input signals generated by one or more sensors of the imprint lithography system, as well as the relationships between these inputs and outputs. Using these multiple inputs and outputs, a matrix of calibration data is generated to produce one or more control parameters that, when applied at various points within the operation of the feedforward-feedback control loop, improve the alignment of the substrate and template by minimizing the vibration caused by the CVA process.

[0033] exist Figure 4 The image shows the use of Figure 1 A block diagram of the feedforward-feedback control loop of the imprint lithography system is provided. An exemplary output of this control loop is a control signal generated to control the movement of the stage in the XYθ direction. This control signal includes one or more parameter values, which are converted into electrical signals and applied to a stage motor (not shown) to move the stage to a desired target position for alignment between the template and the substrate. By continuously monitoring the positional information of the stage and substrate relative to the template and using these measurements, the system described herein successfully reduces vibrations generated by controlled viscosity alignment.

[0034] Figure 4 An example control block diagram is shown. The control system described herein is illustrated as follows: Figure 1This is part of a controller 144 in the system 100 shown. The controller 144 includes at least one central processing unit (CPU) and memory, and can execute instructions stored in the memory to perform one or more of the described operations and / or functions. The controller 144 communicates with one or more memories (e.g., RAM and / or ROM), and in some cases, executes stored instructions to perform one or more control operations. In other cases, the controller 144 may temporarily store data used to calculate and generate the various signals described below in one or more memories. Therefore, the controller 144 controls the system by using computer programs (one or more sets of stored instructions executable by the CPU) and data stored in RAM and / or ROM. Figure 1 System 100. Here, controller 144 may include one or more dedicated hardware or graphics processing units (GPUs) (or may communicate with them) that are different from the CPU, and the GPU or dedicated hardware may perform some of the CPU's processing. Examples of dedicated hardware include application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), and digital signal processors (DSPs). In one embodiment, control system 100 may be implemented as part of controller 144, such as Figure 1 As shown. In some embodiments, controller 144 may be a dedicated controller. In other cases, control system 100 may include multiple controllers communicating with each other and other components of control system 100 to implement the operations described herein.

[0035] The following will describe the execution of control functions according to this disclosure. Figure 4 A block diagram is provided. While various controllers are referenced below, in some embodiments, each controller may include a series of stored instructions executed by the CPU of controller 144 to perform the described functions. In other embodiments, each controller described herein may be embodied as a separate integrated circuit, each having its own CPU and memory, and dedicated to performing the associated processing. In other embodiments, one or more controllers described herein may be embodied as a single integrated circuit. Furthermore, in some embodiments, some of the described controllers may be dedicated processing units and communicate with the CPU of the controller executing the stored instructions to perform the functional operations described herein.

[0036] Figure 4The system includes a sensor 158, a feedforward controller 410, an alignment feedback controller 430 (K1), a stage feedback controller (K2) 450, and a stage amplifier 470 (hereinafter referred to as "amplifier 470"). Positioned between these components are multiple nodes that combine signals by adding, subtracting, or convolving the signals output by the respective controllers that provide the given signals. These components operate as described below to move the stage of the support substrate to a target position representing an alignment error value within a predetermined alignment error range. Each of the alignment feedback controller 430 and the stage feedback controller 450 can be executed as a proportional-integral-derivative (PID) controller or any other feedback controller. The control function executed by the feedback controller is known and does not require further description; it continuously calculates the error value based on the inputs received therein. The first node 420 is located between the feedforward controller 410 and the sensor 158, and generates a first input signal 422 to the alignment feedback controller 430. The first input signal 422 uses a first feedforward control signal 412(f1) representing the alignment reference trajectory and a measurement signal 402(y1) representing the position of a mark on the template relative to a mark on the substrate to represent the feedback control error (e1). The first feedforward signal 412(f1) represents the alignment reference trajectory information and is generated based on the measurement signal 402 and the reference trajectory information stored in memory. Using the first input signal e1, the alignment controller 430 can generate an alignment trajectory (u1) (sometimes referred to as a control command or control effort) that will be used to more quickly align the marks on the substrate and template. The alignment controller 430 outputs alignment trajectory information 432(u1) as input to the second node 440.

[0037] The feedforward controller 410 also outputs a second feedforward control signal 414(f2) to the second node 440. The second feedforward control signal 414(f2) may be the same as the first feedforward signal 412(f1), or it may be a shifted version of the first feedforward signal 412(f1) (e.g., time shift, phase shift, amplitude shift, etc.). In addition to the alignment trajectory information 432 and the second feedforward signal 414, the second node 440 receives stage position information 474(y2) obtained from the stage position sensor 476 of the amplifier 470 as input. The stage position information 474 indicates the current measured position of the stage 106 at the end of a stage movement operation requested by the amplifier 470. These signals are combined to generate a second input signal 442(e2) representing the error rate of the stage feedback controller 450(K2). The stage feedback controller 450(K2) generates a control action representing the stage trajectory information 452(u2) as its output. Although the stage position sensor 476 is shown separately from the stage 106, this is merely exemplary and illustrated for ease of understanding system operation. It should be understood that the stage position sensor 476 may be included within the stage 106. The stage feedback controller 450 continuously calculates error values ​​associated with the position of the stage 106 controlled by the amplifier 470.

[0038] A third node 460 is positioned between the stage feedback controller 450 (K2) and the stage amplifier 470. The third node 460 combines the stage trajectory information 452 (u2) with a third feedforward signal 416 (f3) generated and output by the feedforward controller 410. The third feedforward signal 416 (f3) is a motion control command prediction signal, and the third node 460 combines it with the stage trajectory information 452 (u2) to generate a stage motion control signal 462, which is converted by the stage amplifier 470 into a voltage or current (output control signal 472) for driving the stage 106 in a defined manner.

[0039] Further information defining the aforementioned input and output signals of each of the feedforward controller 410, alignment feedback controller 430, and stage feedback controller 450, as well as the benefits applicable to alignment processing, can be found in the description of U.S. Patent Application No. 16 / 712,739, the entire contents of which are incorporated herein by reference.

[0040] Figure 4An energy controller 401 is also shown, which controls an energy source 138 during controlled viscosity alignment (CVA). During CVA, after the polymerizable material is deposited on the substrate, energy of a predetermined intensity is directed at the substrate for a predetermined duration to increase the viscosity of the polymerizable material during the alignment of various marks in the substrate and the template. In one exemplary operation, multiple energy application periods occur, in which the predetermined intensity of energy is applied for the duration of that period. Figure 5 An exemplary operation of the timing of the CVA process is illustrated. As shown, multiple energy application periods are illustrated as "Slot 1", "Slot 2", "Slot 3", and "Slot 4", where "Slot 1" begins at t1 and has a predetermined duration, "Slot 2" begins at t2 and has a predetermined duration, "Slot 3" begins at t3 and has a predetermined duration, and "Slot 4" begins at t4 and has a predetermined duration. In one embodiment, the durations of the various energy application periods are the same. In another embodiment, some energy application periods have the same duration, while others have different durations. In yet another embodiment, the duration of each successive energy application is shorter than the immediately preceding energy application period. In another embodiment, the duration of each successive energy application is longer than the immediately preceding energy application period. In another embodiment, the spatial distribution of the energy provided varies during the various time slots. In yet another embodiment, additional energy of different wavelengths and intensities is provided after the CVA to further solidify the polymerizable material.

[0041] The advantage of CVA treatment is that by increasing the viscosity of the polymerizable material deposited on the substrate, conventional vibrations associated with alignment processes can be reduced, thereby aligning the marks on the substrate with the marks on the template within an acceptable timeframe. However, the result of CVA treatment is other vibrations that can further affect alignment. For example, the amplitude, frequency, and / or phase of the vibrations introduced into the system may differ each time CVA is applied to a specific area on the substrate. Therefore, it is difficult to correct for these secondary vibrations that may differ in frequency and magnitude after imprinting on continuous areas of the substrate. To compensate for these secondary vibrations, Figure 4 The control system also includes a calibration controller 405, which continuously calculates calibration parameters based on historical input and output values ​​and current real-time input and output values ​​to adjust the learning rate used to generate adaptive vibration cancellation parameters applied to each of the feedforward controller 410, alignment feedback controller 430 and stage feedback controller 450, which will be discussed below.

[0042] The calibration controller 405 generates at least one calibration matrix based on the moving-time window transfer function and covariance. This calibration matrix is ​​a multiple-input multiple-output (MIMO) matrix. The MIMO matrix takes the following signals as inputs, as shown in Table 1:

[0043] ·y1: Position signal 402 sensed by sensor 158, which represents the relative position of the mark on the template with respect to the mark position on the substrate.

[0044] ·y2: Position signal 474 sensed by stage position sensor 476, which indicates the last position of the stage when the most recent alignment action was completed.

[0045] • u1: Control signal output from alignment feedback controller 430, representing alignment trajectory information.

[0046] • u2: Control signal output from platform feedback controller 450, representing platform trajectory information.

[0047] f1: The first feedforward control signal input to the alignment feedback controller 430

[0048] f2: The second feedforward control signal input to the stage feedback controller 450

[0049] f3: The third feedforward control signal input to amplifier 470

[0050] e1: The error value derived from the most recent alignment operation and input to the alignment feedback controller 430.

[0051] e2: The error value derived from the most recent alignment operation and input to the stage feedback controller 450.

[0052] • Represents the historical data values ​​of each of the above inputs.

[0053] Table 1: Input signals for the calibration matrix

[0054] The calibration matrix generated by calibration controller 405 is used to selectively modify the adaptive feedforward learning rate, which is used to generate adaptive functions applied to other controllers 410, 430, and 450, thereby adjusting the parameters of the feedback loop to better correct for vibrations introduced by the CVA process. The adaptive functions are generated using real-time sensing information and offline historical alignment data from previously performed imprint lithography processes. By using multiple input values ​​of various types, calibration controller 405 can improve the adjustment of the adaptive learning rate and the adaptive functions derived using the adjusted adaptive learning rate. The adaptive functions represent vibration (e.g., noise) cancellation signals, which are active cancellation inputs provided to each of the feedforward controller 410, alignment feedback controller 430, and stage feedback controller 450, which actively cancel vibrations caused by the CVA process when generating their respective output signals as described above. The adjustment of the adaptive learning rate and the resulting adaptive function are generated not only using the individual inputs discussed above, but also considering the interactions of the different inputs listed above (such as by calculating the covariance between various calibration controller inputs). More accurate and robust vibration correction can be achieved by considering the interrelationships of one or more calibration controller inputs. Exemplary calibration matrices are shown in Table 2, visualizing the input signals and their relationships. For the various relationships listed below, the calibration matrices calculate the relationships of the moving time windows and the covariance between the signals. Covariance is a measure of the joint variability of two signals.

[0055]

[0056]

[0057] Table 2: Exemplary Calibration Matrix

[0058] For example, the calibration controller 405 generates an initial calibration matrix that includes current real-time inputs and historical input values ​​from previous imprints when CVA processing begins and the viscosity of the polymerizable material increases. Because Figure 4The sampling rate of the control loop is lower than that of a single CVA application, so the control loop can sense and continuously calculate and monitor multiple different inputs, as described above. Once CVA begins, friction increases and vibration decreases. If this is the only vibration, the adaptive function generated and applied to the controller should be reduced. However, since the changes that occur during CVA happen over time and are not uniformly distributed in terms of time, peak frequency, and friction level, the transition from CVA to UV application is not robust, and the alignment error is randomly distributed rather than shrinking to zero. The generated calibration matrix favorably predicts the effects and transitions of CVA processing, so that the adaptive function generated thereby can be used to actively modify how the three feedforward signals (f1, f2, f3) generated by the feedforward controller 410, and how the alignment feedback controller 430 and the stage feedback controller 450 use the respective error rates (e1, e2) input to them along with the three feedforward signals (f1, f2, f3) to improve the generated control signals (u1, u2), thereby actively canceling and correcting the vibration caused by CVA processing. The adaptive function generated from the calibration matrix can include modifications to one or more parameters of the amplitude, phase, and frequency of the control signal used to correct vibrations. The calibration matrix is ​​continuously calculated and updated to monitor the interactions between various matrix input signals to improve the accuracy of the adaptive function generated and applied throughout the system. For example, when calculating the relationship between input and output signals, if their covariance is below a threshold covariance, this indicates that the generated adaptive function reduces the position error during alignment by correcting for vibrations introduced in the CVA process. If the covariance is above the threshold covariance, the calibration controller 405 modifies the learning rate used to generate the adaptive function by reducing the weights associated with one or more input signals and / or identifying other input signals to be used as part of the learning rate, in an attempt to minimize the covariance value between the input signals being used. In an exemplary illustration, if the control action used to minimize vibrations from the CVA process increases but the error rate also increases, the covariance will also increase. This indicates that the control action is too large. Therefore, the corrections made are used to modify the control gain and / or learning so that the resulting output signal reduces the control action.

[0059] Turn now Figure 6 , Figure 6 An exemplary control algorithm applied to feedforward alignment control by a calibration controller 405 according to this disclosure is shown. The following description will utilize... Figures 1 to 4Associated reference numerals indicate the processing unit performing algorithmic control. In step S602, the calibration controller receives multiple input signals representing system input control signals and system output control signals. These system input control signals and system output control signals include signals used as inputs and outputs to the feedforward controller 410, alignment feedback controller 430, and stage feedback controller 450. Other inputs received in S602 include sensed position inputs from one or more sensors in the imprint lithography system (including, but not limited to, moiré sensors and stage position sensors). For example, the signals received in S602 include at least the signals listed in Table 1 above. These input signals received in S602 are continuously received according to the sampling rate of the feedforward-feedback alignment control system. In another embodiment, the input signals received in S602 include historical input signals. These historical input signals are of the same type as discussed above, but note the difference that these historical input signals are derived from previously completed imprinting processes.

[0060] In step S604, the calibration controller uses the inputs received in S602 and system model parameters to generate a calibration matrix, the system model parameters defining which transfer functions are applied throughout the control system. When generating the calibration matrix in S604, the calibration controller decouples, filters, and normalizes the inputs received in S602, allowing the relationship between one or more received inputs to be determined. By determining the relationship between one or more input signals, the learning rate used to generate the adaptive functions to be applied to various system components is improved, facilitating faster alignment of marks on the substrate and template. An exemplary calibration matrix includes a transfer function matrix as shown in Table 3, which illustrates the relationship between various input signals.

[0061]

[0062] u1 = K1g1 = K1(f1 - y1)

[0063] N2=K2e2=K2(f2-y2)

[0064]

[0065] Table 3: Transfer Function Matrix

[0066] in,

[0067] K1 indicates alignment feedback controller 430

[0068] K2 indicates stage feedback controller 450

[0069] ·y1: Position signal 402 sensed by sensor 158, which represents the relative position of the mark on the template with respect to the mark position on the substrate.

[0070] ·y2: Position signal 474 sensed by stage position sensor 476, which indicates the last position of the stage when the most recent alignment action was completed.

[0071] • u1: Control signal output from alignment feedback controller 430, representing alignment trajectory information.

[0072] • u2: Control signal output from platform feedback controller 450, representing platform trajectory information.

[0073] f1: The first feedforward control signal input to the alignment feedback controller 430

[0074] f2: The second feedforward control signal input to the stage feedback controller 450

[0075] f3: The third feedforward control signal input to amplifier 470

[0076] The time-varying transfer function g was used when defining the relationship between y1, u1, and f2. 11 and g 12 As used in this article, g 11 (t) represents the time-varying transfer function between the input u1+f2 to the stage feedback controller 450 and the output y1 representing the position signal 402 sensed by the sensor 158. Furthermore, g 12 (t) represents the time-varying transfer function between the input u2+f3 of the stage amplifier 470 and the output y1 of the position signal 402 sensed by the sensor 158. The time-varying transfer function g is used in defining the relationship between y2 and u2 and f3. 21 and g 22 As used in this article, g 21 (t) represents the time-varying transfer function between the input u1+f2 of the stage feedback controller 450 and the output y2 of the stage position sensor 476. Furthermore, g 22 (t) represents the time-varying transfer function between the input u2+f3 of the stage amplifier 470 and the output y2 of the stage position sensor 476. Furthermore, when defining the relationship between the feedforward control signals f1, f2, and f3 and the position signals y1 and y2, the calibration matrix considers the control parameters of the feedforward controller 410, where ψ 11 It is a feedforward controller based on y1 and f1; ψ 12 This represents a feedforward controller based on y2 and f1; ψ 21 This represents a feedforward controller based on y1 and f2; ψ 22This represents a feedforward controller based on y2 and f2; ψ 31 This represents a feedforward controller based on y1 and f3; ψ 32 This represents a feedforward controller based on y2 and f3.

[0077] The calibration matrix generated in S604 also includes a covariance matrix that identifies the variance ratio between the input signals. Example covariance matrices are shown in Table 4 below:

[0078] cov(∈, Y)

[0079] =E[(∈-E[∈])(yE[Y])]

[0080]

[0081] Table 4: Covariance Matrix

[0082] in,

[0083] ·y1: Position signal 402 sensed by sensor 158, which represents the relative position of the mark on the template with respect to the mark position on the substrate.

[0084] ·y2: Position signal 474 sensed by stage position sensor 476, which indicates the last position of the stage when the most recent alignment action was completed.

[0085] f3: The third feedforward control signal input to amplifier 470

[0086] e1: The error value derived from the most recent alignment operation and input to the alignment feedback controller 430.

[0087] e2: The error value derived from the most recent alignment operation and input to the stage feedback controller 450.

[0088] • E[]: is the expected value of any object within the brackets. This can be implemented as a weighted average calculated over a moving time window.

[0089] S604 also includes decoupling these input signals, enabling the system to construct parameters applicable to each system controller to achieve the desired effect of correcting vibrations introduced during CVA processing. When the transfer function matrices in Table 3 are decoupled, we obtain the following results:

[0090] y1 = G1(K1e1)

[0091] y2=G2(K2e2)

[0092] Among them, G i It is a transfer function, which represents the detected error (e) i ) and the sensed position signal (yi The relationship between ).

[0093] Furthermore, when the covariance matrix is ​​decoupled, we obtain the following results.

[0094] cov(e1, u1)

[0095] cov(e2, u2)

[0096] cov(f3, e2)

[0097] In doing so, we obtain the covariance between the alignment control feedback error e1 and the control signal u1 generated by the alignment feedback controller 430, the covariance between the stage control feedback error e2 and the control signal u2 generated by the alignment feedback controller 430, and the covariance between the feedforward control signal f3 and the stage control feedback error e2.

[0098] In step S606, the calibration controller 405 adjusts the learning rate associated with each system component. In operation, the decoupling covariance matrix is ​​used to adjust the learning rate for generating an adaptive function, which will be applied to the system components to provide control parameters for incorporating active vibration cancellation into the alignment process utilizing CVA alignment processing. The learning rate of the adaptive function to be applied to the alignment feedback controller 430 is adjusted according to the following equation:

[0099] ρ K1 =ρ K1 -β1{cov(e1, u1)}

[0100] Where, ρ K1 Let β1 represent the learning rate of the adaptive function applied to K1, and let β1 represent the learning rate adjustment constant or function. The learning rate of the adaptive function applied to the stage feedback controller is adjusted according to the following equation:

[0101] ρ K2 =ρ K2 -β2{cov(e2, u2)}

[0102] Where, ρ K2 Let β2 represent the learning rate of the adaptive function applied to K2, and let β2 represent the learning rate adjustment constant or function. Finally, the learning rate of the adaptive function to be applied to the feedforward controller 410 is adjusted according to the following equation:

[0103] ρ 32 =ρ 32 -β 32 {cov(f3, e2)}

[0104] Where ρ 32β represents the learning rate of the adaptive function with the feedforward controller 410 as the object. 32 This represents the learning rate adjustment constant or function.

[0105] In step S608, using the learning rate generated in S606, a component-specific adaptive function can be generated. The first adaptive function is applied to the feedforward controller 410 generated using the decoupling transfer function matrix in Table 3, according to the following equation:

[0106] Δψ 32 =ρ 32 Ψ 32 (f3, e2)

[0107] Where, Δψ 32 ρ represents the change in the applied control parameter. 32 Ψ represents the learning rate of the adaptive function. 32 Let represent the adaptive function of the feedforward controller 410. A second adaptive function is generated according to the following equation and applied to the alignment feedback controller 430:

[0108] ΔK1=ρ K1 Θ1(e1, u1)

[0109] Where ΔK1 represents the change applied to the alignment feedback controller 430, ρ K1 Let Θ1 represent the learning rate applied to controller 430, and Θ1 represent the adaptive function of K1. A third adaptive function is generated according to the following equation and applied to the stage feedback controller 450:

[0110] ΔK2=ρ K2 Θ2(e2, u2)

[0111] Where ΔK2 represents the change applied to the stage feedback controller 450, ρ K2 Θ2 represents the learning rate applied to the stage feedback controller 450, and Θ2 represents the adaptive function of K2.

[0112] Back Figure 4In S606, at each sampling time according to the sampling rate of the system described herein, the learning rate is continuously adjusted to continuously update component-specific adaptive functions, which positively contribute to eliminating noise introduced during the controlled viscosity alignment process. These updated component-specific adaptive functions are applied in S610. At each sampling time, a first adaptive function 406 is applied to the feedforward controller 410 and used by the feedforward controller 410 in generating the first feedforward control signal f1, the second feedforward control signal f2, and the third feedforward control signal f3. A second adaptive function is applied to the alignment feedback controller 430 and is thus used to generate the control signal u1 in addition to being used as its regular input as described above. A third adaptive function is applied to the stage feedback controller 450 and is thus used to generate the control signal u2 in addition to being used as its regular input.

[0113] In step S610, an output control signal 472 is provided by an amplifier 470, which controls the movable stage to approach the target position based on the generated output signal. In step S612, a sensor 158 determines whether the error value associated with the relative position of the marks on the substrate and the marks on the template is within a predetermined error range. If the determination in S612 is positive (indicating that the error value is within an acceptable range), the determination indicates that the marks on the template and the marks on the substrate are aligned with each other, and the control algorithm ends in S615.

[0114] Therefore, the control algorithms detailed above can be used in semiconductor manufacturing processes for manufacturing one or more articles or devices. These processes, including but not limited to: imprint lithography; photolithography; baking; oxidation; layer formation; deposition; doping; etching; dross removal; dicing; bonding; and encapsulation, can be performed on substrates that have been successfully aligned according to the control algorithms described above. Other known article manufacturing steps and processes can also be used to process the substrate, including, for example, inspection, curing, oxidation, layer formation, deposition, doping, planarization, etching, formable material removal, dicing, bonding, and encapsulation. Based on the above, the substrate can be processed to produce multiple articles (devices).

[0115] In one example, the process may include dispensing an imprinting resist (e.g., a liquid) onto a substrate and bringing the imprinting resist into contact with an object, such that the object, which has a pattern thereon, is in contact with the imprinting resist. The process includes an alignment process of aligning the substrate and the object (e.g., a stencil) to predetermined alignment positions, followed by processing the substrate with the imprinting resist dispensed to manufacture an article. In this process, energy is applied to the substrate to cure the resist and form a pattern on the substrate corresponding to the pattern on the stencil. This process is repeated such that the object and substrate are aligned before the applied energy cures the resist.

[0116] Several implementations have been described. However, it is understood that various modifications can be made without departing from the spirit and scope of this disclosure. Therefore, other implementations are also within the scope of the following claims.

[0117] Embodiments of this disclosure can be implemented by providing a program that implements one or more functions of the above embodiments to a system or device via a network or storage medium, and by using one or more processors of the computer of the system or device to read and execute the program. Alternatively, embodiments of this disclosure can be implemented by circuitry (e.g., application-specific integrated circuits (ASICs)) that implement one or more functions.

[0118] Embodiments of the invention can also be implemented by a computer that reads and executes computer-executable instructions (e.g., one or more programs) recorded on a storage medium (also more fully referred to as a "non-transitory computer-readable storage medium") to perform one or more functions in the above embodiments, and / or includes one or more circuits (e.g., application-specific integrated circuits (ASICs)) for performing one or more functions in the above embodiments. Furthermore, embodiments of the invention can be implemented using a method by which the computer of the system or device, for example, reads and executes the computer-executable instructions from the storage medium to perform one or more functions in the above embodiments, and / or controls the one or more circuits to perform one or more functions in the above embodiments. The computer may include one or more processors (e.g., central processing unit (CPU), microprocessor unit (MPU)) and may include separate computers or a network of separate processors to read and execute the computer-executable instructions. The computer-executable instructions may be provided to the computer, for example, from a network or the storage medium. The storage medium may include, for example, a hard disk, random access memory (RAM), read-only memory (ROM), the memory of a distributed computing system, or an optical disc (such as a compact disc (CD), a digital versatile optical disc (DVD), or a Blu-ray disc (BD)). TM One or more of the following: flash memory devices and memory cards.

[0119] In the description, specific details are set forth in order to provide a thorough understanding of the disclosed examples. In other instances, well-known methods, processes, components, and circuits are not described in detail to avoid unnecessarily lengthening this disclosure.

[0120] It should be understood that if an element or part is referred to herein as being "on," "against," "connected to," or "attached to" another element or part, then the element or part may be directly on, directly abutting, directly connected to, or directly attached to the other element or part, or there may be intervening elements or parts. Conversely, if an element is referred to as being "directly on," "directly connected to," or "directly attached to" another element or part, then there are no intervening elements or parts. If provided as such, the term "and / or" as used includes any and all combinations of one or more of the associated listed items.

[0121] For ease of description, spatial relative terms such as “below,” “under,” “below,” “down,” “above,” “up,” “proximal,” and “farthest” are used herein to describe the relationship between one element or feature and another element(s) as shown in the figures. However, it should be understood that, in addition to the orientation depicted in the figures, spatial relative terms are also intended to cover different orientations of the device during use or operation. For example, if the device in the figure is flipped, an element described as “below” or “under” other elements or features will be oriented as “above” other elements or features. Thus, spatial relative terms such as “below” can cover both above and below orientations. The device may be oriented in other ways (rotated 90° or in other orientations), and the spatial relative descriptors used herein will be explained accordingly. Similarly, where applicable, spatial relative terms such as “proximal” and “farthest” may be used interchangeably.

[0122] As used herein, the term "about" means, for example, within 10%, within 5%, or less. In some embodiments, the term "about" may refer to within measurement error.

[0123] In this document, the terms first, second, third, etc., may be used to describe various elements, components, regions, parts, and / or intervals. It should be understood that these elements, components, regions, parts, and / or intervals should not be limited by these terms. These terms are merely used to distinguish one element, component, region, part, or interval from another. Therefore, without departing from the teachings herein, the first element, component, region, part, or interval discussed below may be referred to as the second element, component, region, part, or interval.

[0124] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. As used herein, unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well. It should also be understood that, when used in this specification, the terms “comprising” and / or “including” indicate the presence of the stated features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof not explicitly stated.

[0125] The foregoing only illustrates the principles of this disclosure. Various variations and modifications to the described exemplary embodiments will be apparent to those skilled in the art in light of the teachings herein.

[0126] In describing the exemplary embodiments shown in the accompanying drawings, specific terminology has been used for clarity. However, this disclosure is not intended to limit us to the specific terminology chosen, and it should be understood that each particular element includes all technical equivalents that operate in a similar manner.

Claims

1. A method for imprinting a substrate supported on a movable stage, the method comprising: The liquid resist on the substrate comes into contact with the template; Using the first set of control parameters, feedback-feedforward control processing is used to align the substrate with the template. Obtain calibration data; Increase the viscosity of at least a portion of the resist; After increasing the viscosity of the resist, the substrate and the template are aligned using the feedback-feedforward control process with a second set of control parameters, wherein the second set of control parameters is determined based on the calibration data; and The resist beneath the template is cured. Specifically, while aligning the substrate and the template using the first set of control parameters and the feedback-feedforward control processing, the calibration data is generated. The method further includes generating the calibration data by the following operations: Receive a set of alignment controls that change over time; The set of alignment controls is converted into a set of alignment state values ​​that change with the first state variable; Receive a set of location estimates that change over time; The set of position estimates is converted into a set of position state values ​​that change with the first state variable; and Generate one or more covariance values. The generation of the calibration data also includes obtaining a second set of alignment controls and a second set of position estimates, wherein the second set of alignment controls and the second set of position estimates are based on the previously performed imprinting process.

2. The method according to claim 1, wherein, The set of alignment controls includes: One or more adaptive functions used in the feedback-feedforward control process; One or more output control signals used to control the position of the stage; and An energy controller signal that controls one or more parameters of the energy applied to increase the viscosity of the resist.

3. The method according to claim 1, wherein, The set of position estimates includes: a first position signal indicating the position of the mark on the template relative to the mark on the substrate; and a second position signal indicating the position of the stage.

4. The method according to claim 1, wherein, The generated covariance values ​​are used to generate the second set of control parameters by modifying one or more control functions applied during the feedback-feedforward control process.

5. The method according to claim 1, wherein, The first set of control parameters includes frequency components, the second set of control parameters represents the updated frequency components generated based on the calibration data, and the method further includes: The substrate and the template are aligned based on the updated frequency components in the second set of control parameters.

6. An imprint lithography system for controlling the alignment of an imprint lithography template relative to a substrate, the imprint lithography system comprising: A stage, configured to hold the substrate and be movable, allowing the position of the stage to be modified; A depositor configured to deposit a resist on the substrate; An energy controller is configured to apply energy to the resist to increase the viscosity of the resist; A sensor is configured to sense the position of the substrate relative to the imprint lithography template; as well as At least one controller communicating with the stage and the sensor, the at least one controller being configured to perform the following operations based on the contact between the substrate having a liquid imprint resist and the template: Using the first set of control parameters, feedback-feedforward control processing is used to align the substrate with the template. Obtain calibration data; Increase the viscosity of at least a portion of the resist; After increasing the viscosity of the resist, the substrate and the template are aligned using the feedback-feedforward control process with a second set of control parameters, wherein the second set of control parameters is determined based on the calibration data; and The resist beneath the template is cured. Meanwhile, the at least one controller also generates the calibration data while aligning the substrate and the template using the first set of control parameters and the feedback-feedforward control processing. The at least one controller further generates the calibration data through the following operations: Receive a set of alignment controls that change over time; The set of alignment controls is converted into a set of alignment state values ​​that change with the first state variable; Receive a set of location estimates that change over time; The set of position estimates is converted into a set of position state values ​​that change with the first state variable; and Generate one or more covariance values. The generation of the calibration data also includes obtaining a second set of alignment controls and a second set of position estimates, wherein the second set of alignment controls and the second set of position estimates are based on the previously performed imprinting process.

7. The imprint lithography system according to claim 6, wherein, The set of alignment controls includes: One or more adaptive functions used in the feedback-feedforward control process; One or more output control signals used to control the position of the stage; and An energy controller signal that controls one or more parameters of the energy applied to increase the viscosity of the resist.

8. The imprint lithography system according to claim 6, wherein, The set of position estimates includes: a first position signal indicating the position of the mark on the template relative to the mark on the substrate; and a second position signal indicating the position of the stage.

9. The imprint lithography system according to claim 6, wherein, The generated covariance values ​​are used to generate the second set of control parameters by modifying one or more control functions applied during the feedback-feedforward control process.

10. The imprint lithography system according to claim 6, wherein, The first set of control parameters includes a frequency component, the second set of control parameters represents an updated frequency component generated based on the calibration data, and the at least one controller also aligns the substrate with the template based on the updated frequency component in the second set of control parameters.

11. A method of manufacturing an article, comprising imprinting a substrate supported on a movable platform using the method of claim 1, the method further comprising: The liquid resist is applied onto the substrate; The liquid resist is brought into contact with the template, and the template has a pattern that is in contact with the liquid resist; The substrate, to which the liquid resist is dispensed, is processed to manufacture an article.

12. The method for manufacturing an article according to claim 11, wherein, The processing of the substrate further includes: Energy is applied to the substrate to cure the liquid resist and form a pattern on the substrate that corresponds to the pattern on the template. The imprinting method is performed repeatedly while the template is in contact with the liquid resist, so that the template is aligned with the substrate before the resist is cured by applied energy.

Images