Subfield control for lithographic processes and associated apparatus

Abstract

A method for controlling a lithographic apparatus configured to pattern an exposure field on a substrate comprising at least a subfield is disclosed, the method comprising: acquiring an initial spatial distribution, the initial spatial distribution being associated with a spatial variation in a performance parameter, the spatial variation of the performance parameter is associated on the substrate across at least a first layer of a sub-field of the exposure field; and decomposing the initial spatial distribution into at least a first component spatial distribution for controlling the lithographic apparatus at a first spatial scale and a second component spatial distribution for controlling the lithographic apparatus at a second spatial scale associated with the size of the subfield, wherein the decomposition comprises co-optimizing the first component spatial distribution and the second component spatial distribution based on correcting spatial variations in the performance parameters across the subfields.

Description

[0001] Description of the case

[0002] This application is a divisional application of the Chinese invention patent application with application number 202080027012.7 filed on March 5, 2020 and entitled “Sub-field control and related equipment for lithography processes”.

[0003] CROSS-REFERENCE TO RELATED APPLICATIONS

[0004] This application claims priority to EP application 19167211.2 filed on April 4, 2019, EP application 19172479.8 filed on May 3, 2019, EP application 19177106.2 filed on May 28, 2019, and EP application 19192433.1 filed on August 20, 2019, the entire contents of which are incorporated herein by reference. Technical Field

[0005] The present invention relates to methods and apparatus for applying a pattern to a substrate and / or measuring such a pattern in a lithographic process. Background Art

[0006] A lithographic apparatus is a machine that applies a desired pattern to a substrate, usually to a target portion of the substrate. A lithographic apparatus can be used, for example, to manufacture an integrated circuit (IC). In this case, a patterning device (alternatively referred to as a mask or reticle) can be used to generate a circuit pattern to be formed on a separate layer of the IC. The pattern can be transferred to a target portion (e.g., including a portion, one or several dies) on a substrate (e.g., a silicon wafer). The transfer of the pattern is usually performed via imaging onto a layer of radiation-sensitive material (resist) provided on the substrate. Typically, a single substrate will contain a network of adjacent target portions that are continuously patterned. Known lithographic apparatus include: a so-called stepper, in which each target portion is irradiated by exposing the entire pattern to the target portion at once; and a so-called scanner, in which each target portion is irradiated by scanning the pattern by a radiation beam in a given direction (the "scanning" direction), while scanning the substrate synchronously parallel to or antiparallel to the direction. The pattern can also be transferred from the patterning device to the substrate by imprinting the pattern onto the substrate.

[0007] In order to monitor the lithography process, parameters of the patterned substrate are measured. Parameters may include, for example, the overlay error between successive layers formed in or on the patterned substrate and the critical line width (CD) of the developed photoresist. The measurement may be performed on a product substrate and / or a dedicated measurement target. There are a variety of techniques that can be used to measure the microstructures formed in the lithography process, including the use of scanning electron microscopes and various special tools. A fast and non-invasive form of special inspection tool is a scatterometer, in which a radiation beam is directed onto a target on the surface of the substrate and the properties of the scattered or reflected beam are measured. Two main types of scatterometers are known. Spectral scatterometers direct a broadband radiation beam onto a substrate and measure the spectrum (intensity as a function of wavelength) of the radiation scattered into a specific narrow angular range. Angular resolved scatterometers use a monochromatic radiation beam and measure the intensity of the scattered radiation as a function of angle.

[0008] Examples of known scatterometers include angle-resolved scatterometers of the type described in US2006033921A1 and US2010201963A1. Such scatterometers use relatively large targets, such as a 40 μm×40 μm grating, and the measurement beam generates a spot that is smaller than the grating (i.e., the grating is underfilled). In addition to measuring feature shapes by reconstruction, diffraction-based overlay can also be measured using such a device, as described in published patent application US2006066855A1. Diffraction-based overlay measurements using dark field imaging of diffraction orders can make overlay measurements of smaller targets. Examples of dark field imaging measurements can be found in international patent applications WO2009 / 078708 and WO2009 / 106279, which are incorporated herein by reference in their entirety. Further developments of this technology have been described in published patent publications US20110027704A, US20110043791A, US2011102753A1, US20120044470A, US20120123581A, US20130258310A, US20130271740A and WO2013178422A1. These targets can be smaller than the illumination spot and can be surrounded by product structures on the wafer. Multiple gratings can be measured in one image using a composite grating target. The contents of all these applications are also incorporated herein by reference.

[0009] At present, the overlay error is controlled and corrected by the correction model described in US2013230797A1, for example. Advanced process control techniques have been introduced in recent years, and these process control techniques use the measurement of the measurement target applied to the substrate and the device pattern applied. These targets allow the overlap to be measured using high-throughput inspection equipment (such as a scatterometer), and the measurement can be used to generate corrections, which are fed back into the lithography equipment when the subsequent substrate is patterned. Examples of advanced process control (APC) are described in US2012008127A1, for example. The inspection equipment can be separated from the lithography equipment. In the lithography equipment, the wafer correction model is usually applied based on the measurement of the overlap target provided on the substrate, and the measurement is used as a preparatory step for each patterning operation. The current correction model includes a high-order model to correct the nonlinear distortion of the wafer. The correction model can also be extended to take into account other measured and / or calculated effects, such as thermal deformation during the patterning operation.

[0010] While using higher order models may be able to account for more effects, such models may be of limited use if the patterning device itself does not provide control over the corresponding parameters during the patterning operation. Furthermore, even advanced correction models may not be sufficient or optimized to correct certain overlay errors.

[0011] It would be desirable to improve such process control methods.

[0012] In addition to the control of higher order overlay errors, it is also necessary to verify that the control profile used by the lithographic apparatus does not induce any undesired dynamic effects that could degrade pattern fidelity (e.g. geometric accuracy).An object of the present invention is to provide additional control methods aimed at ensuring the geometric integrity of patterning operations. Summary of the invention

[0013] In a first aspect of the present invention, a method for controlling a lithography process for exposing a pattern on an exposure field is provided, the method comprising: obtaining an initial spatial distribution associated with a spatial variation of a performance parameter of a first layer for a subfield across the exposure field; decomposing the initial spatial distribution into at least a first component spatial distribution for controlling a lithography device to correct the spatial variation of the performance parameter and a second component spatial distribution for controlling the lithography device or another device to correct the spatial variation of the performance parameter; and jointly optimizing the first component spatial distribution and the second component spatial distribution to achieve better correction of the spatial variation of the performance parameter compared to correction based on controlling the spatial variation of the performance parameter of the lithography device using the initial spatial distribution.

[0014] In a second aspect of the invention there is provided a computer program comprising program instructions operable when run on a suitable apparatus to perform the method of the first aspect.

[0015] In a third aspect of the present invention, a method for determining an overlap correction distribution is provided, the method comprising: obtaining an overlap error distribution between a first layer and a second layer across subfields within an exposure field; and determining a first overlap correction distribution for the first layer and a second overlap correction distribution for the second layer, wherein the first correction distribution and the second correction distribution are added to mitigate the overlap error distribution across subfields and are jointly optimized to achieve desired imaging quality across both the first layer and the second layer.

[0016] In a fourth aspect of the present invention, a method for determining overlay error correction is provided, the method comprising: obtaining an overlay error distribution extending across multiple subfields along a scanning direction; and determining a mask correction across at least two adjacent subfields in the multiple subfields, wherein the mask correction is based on an expected improvement in the continuity of the overlay error distribution between the above-mentioned adjacent subfields.

[0017] In a fifth aspect of the present invention, a method for providing APC control input to an APC controller and providing subfield control input to a lithography device is provided, the method comprising: a) obtaining a fingerprint of parameters measured across a substrate; b) decomposing the fingerprint into a global component and a subfield component; c) using the global component and the subfield component and the control characteristics of the lithography device to determine a common optimization control strategy; and d) determining the APC control input and the subfield control input based on the common optimization control strategy.

[0018] In a sixth aspect of the present invention, a method for determining overlay error correction is disclosed, the method comprising: obtaining an overlay error distribution extending across multiple subfields; determining a mask correction across at least two adjacent subfields in the multiple subfields, wherein the mask correction is based on an expected improvement in the continuity of the overlay error distribution between the adjacent subfields.

[0019] In a seventh aspect of the present invention, a method for jointly optimizing global and subfield control strategies for a lithography process is disclosed, comprising a) acquiring global measurement data and in-die measurement data, b) decomposing the in-die measurement data into global component data and subfield component data, c) determining a jointly optimized control strategy using the global component data, the subfield component data and known scanner control characteristics, and d) providing a global control input and a subfield control input based on the jointly optimized control strategy.

[0020] In an eighth aspect of the present invention, a method for configuring a process associated with manufacturing a semiconductor device on a substrate is provided, the method comprising: determining an expected variation in the ratio between dimensions of features provided to the substrate using a patterning process; and configuring the patterning process and / or a patterning device used in the patterning process to at least partially correct the expected variation in the ratio across regions on the substrate.

[0021] In a ninth aspect of the present invention, there is provided a method for determining a sampling scheme associated with an exposure field on a substrate undergoing a patterning process, the method comprising: deriving an expected distribution of pattern size errors and / or pattern positioning errors across the exposure field due to the time dependence of synchronization errors between the movement of a patterning device and a substrate during the patterning process; and determining the sampling scheme based on the expected accuracy of characterizing the expected distribution using measurements of pattern size and / or pattern position according to the sampling scheme.

[0022] In a tenth aspect of the present invention, a method for controlling a lithographic device is provided, the lithographic device being configured to pattern an exposure field on a substrate comprising at least subfields, the method comprising: obtaining an initial spatial distribution, the initial spatial distribution being associated with a spatial variation of a performance parameter, the spatial variation of the performance parameter being associated with a first layer of subfields at least across the exposure field on the substrate; and decomposing the initial spatial distribution into at least a first component spatial distribution for controlling the lithographic device at a first spatial scale and a second component spatial distribution for controlling the lithographic device at a second spatial scale associated with the size of the subfield, wherein the decomposition comprises: jointly optimizing the first component spatial distribution and the second component spatial distribution based on correcting the spatial variation of the performance parameter across the subfields.

[0023] Other aspects, features and advantages of the present invention, as well as the structure and operation of various embodiments of the present invention are described in detail below with reference to the accompanying drawings. Note that the present invention is not limited to the specific embodiments described herein. Such embodiments are presented herein for illustrative purposes only. Based on the teachings contained herein, other embodiments will be clear to those skilled in the relevant art. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Embodiments of the present invention will now be described by way of example with reference to the accompanying drawings, in which:

[0025] Figure 1 depicts a lithographic apparatus that, together with other equipment, forms a production facility for semiconductor devices;

[0026] Figure 2 A schematic diagram depicting overall lithography represents the collaboration between three key technologies to optimize semiconductor manufacturing;

[0027] Figure 3 Exemplary sources of processing parameters are shown;

[0028] Figure 4 is a graph of overlap versus field position showing measured overlap; and a fit correction modeled and / or implemented using a control grid determined by a method according to an embodiment of the present invention;

[0029] Figure 5Conceptually illustrating two examples of control distribution decomposition according to the first main embodiment of the present invention;

[0030] Figure 6 is a flowchart describing a method according to an embodiment of the present invention;

[0031] Figure 7 is a flowchart describing a method according to an embodiment of the present invention;

[0032] Figure 8 conceptually illustrates a decomposition of the control distribution according to the second main embodiment of the present invention;

[0033] Fig. 9 is a flowchart describing a method according to an embodiment of the present invention;

[0034] Fig.10 Conceptually describing subfield-specific reticle correction configured to reduce discontinuities in overlay error across boundaries between subfields in accordance with embodiments of the present invention;

[0035] Fig.11 is a flow chart depicting co-optimized global and sub-field control according to an embodiment of the present invention;

[0036] Fig.12 (a) to Fig.12 (c) depicts an example of how global and intra-die fingerprint components may be assigned to global or sub-field control infrastructure.

[0037] Fig.13 The overlap control distribution and its associated MSD (fading) and MA (overlap error) are shown.

[0038] Fig.14 A method for correcting the effect of non-constant decay within an exposure field on the pattern fidelity of a contact hole feature according to an embodiment is shown.

[0039] 15(a) to 15(c) depict examples of distributing subfields across a reticle between two reticles.

[0040] Fig.16 Control distributions associated with the subfield distributions shown in FIG. 15( a ) to FIG. 15 ( c ) are depicted.

[0041] Fig.17 An embodiment is depicted in which the reticle is moved between two separate exposure steps. DETAILED DESCRIPTION

[0042] Before describing embodiments of the present invention in detail, it is helpful to present an example environment in which embodiments of the present invention may be implemented.

[0043] Figure 1 At 200, a lithographic apparatus LA is shown as part of an industrial production facility implementing a high-volume lithographic manufacturing process. In this example, the manufacturing process is applicable to the manufacture of semiconductor products (integrated circuits) on substrates such as semiconductor wafers. Those skilled in the art will appreciate that a variety of products can be manufactured by processing different types of substrates in variations of the process. The production of semiconductor products is used purely as an example that has great commercial significance today.

[0044] In a lithographic apparatus (or simply "lithography tool" 200), a measurement station MEA is shown at 202 and an exposure station EXP is shown at 204. A control unit LACU is shown at 206. In this example, each substrate visits a measurement station and an exposure station to have a pattern applied. For example, in an optical lithographic apparatus, a projection system is used to transfer a product pattern from a patterning device MA to a substrate using conditioned radiation and a projection system. This is done by forming an image of the pattern in a layer of radiation-sensitive resist material.

[0045] The term "projection system" as used herein should be broadly interpreted to cover any type of projection system, including refractive, reflective, catadioptric, magnetic, electromagnetic and electrostatic optical systems, or any combination thereof, which projection system is suitable for the exposure radiation used or other factors (such as the use of immersion liquid or the use of vacuum). The patterned MA device can be a mask or a reticle that imparts a pattern to a radiation beam transmitted or reflected by the patterning device. Known operating modes include stepping mode and scanning mode. It is known that the projection system can cooperate with the support and positioning system for the substrate and the patterning device in a variety of ways to apply the desired pattern to many target portions across the substrate. A programmable patterning device can be used instead of a reticle with a fixed pattern. For example, the radiation can include electromagnetic radiation in the deep ultraviolet (DUV) or extreme ultraviolet (EUV) band. The present disclosure is also applicable to other types of lithography processes, such as imprint lithography and direct write lithography, for example by electron beam.

[0046] The lithographic apparatus control unit LACU controls all movements and measurements of the various actuators and sensors to receive the substrate W and mask MA and implement the patterning operation. The LACU also includes signal processing and data processing capabilities to implement the desired calculations related to the operation of the apparatus. In practice, the control unit LACU will be implemented as a system consisting of many sub-units, each of which handles real-time data acquisition, processing and control of a subsystem or component within the apparatus.

[0047] Before the pattern is applied to the substrate at the exposure station EXP, the substrate is processed at the measurement station MEA so that various preparation steps can be performed. The preparation steps may include mapping the surface height of the substrate using a level sensor and measuring the position of the alignment marks on the substrate using an alignment sensor. The alignment marks are nominally arranged in a regular grid pattern. However, due to inaccuracies in creating the marks and due to deformations of the substrate during its entire processing, the marks deviate from the ideal grid. Therefore, in addition to measuring the position and orientation of the substrate, if the device is to print product features in the correct position with very high accuracy, the alignment sensor must measure the positions of many marks across the substrate area in detail in practice. The device can be a so-called dual-stage type, which has two substrate stages, each substrate stage having a positioning system controlled by a control unit LACU. When one substrate on one substrate stage is exposed at the exposure station EXP, the other substrate can be loaded onto the other substrate stage at the measurement station MEA, so that various preparation steps can be performed. Therefore, the measurement of the alignment marks is very time-consuming, and providing two substrate stages can significantly increase the output of the device. If the position sensor IF is not able to measure the position of the substrate table when it is located at the measurement station and the exposure station, a second position sensor may be provided to enable the position of the substrate table to be tracked at both stations. The lithographic apparatus LA may, for example, be of a so-called dual-stage type having two substrate tables and two stations (an exposure station and a measurement station) between which the substrate table can be exchanged.

[0048] Within a production facility, the apparatus 200 forms part of a "lithography cell" or "lithography cluster" which also includes a coating apparatus 208 for applying photoresist and other coatings to a substrate W for patterning by the apparatus 200. On the output side of the apparatus 200, a baking apparatus 210 and a developing apparatus 212 are provided for developing the exposed pattern into a physical resist pattern. Between all these apparatuses, a substrate handling system is responsible for supporting the substrates and transferring them from one apparatus to the next. These apparatuses, which are generally collectively referred to as tracks, are controlled by a track control unit, which itself is controlled by a supervisory control system SCS, which also controls the lithography apparatus via the lithography apparatus control unit LACU. Thus, the different apparatuses can be operated to maximize yield and processing efficiency. The supervisory control system SCS receives recipe information R which provides a detailed definition of the steps to be performed to create each patterned substrate.

[0049] Once the pattern is applied and developed in the lithography unit, the patterned substrate 220 is transferred to other processing equipment, such as shown in 222, 224, and 226. Various processing steps are implemented in various equipment in a typical production facility. For example, the equipment 222 in this embodiment is an etching station, and the equipment 224 performs a post-etching annealing step. Other physical and / or chemical processing steps are applied in other equipment 226, etc. Manufacturing a real device may require multiple types of operations, such as the deposition of materials, the modification of surface material properties (oxidation, doping, ion implantation, etc.), chemical mechanical polishing (CMP), etc. In fact, the equipment 226 can represent a series of different processing steps performed in one or more devices. As another example, equipment and processing steps for realizing self-aligned multiple patterning can be provided to produce multiple smaller features based on the precursor pattern placed by the lithography equipment.

[0050] As is known, the manufacture of semiconductor devices involves multiple repetitions of such processing to build up device structures with appropriate materials and patterns layer by layer on a substrate. Thus, the substrates 230 arriving at the lithography cluster may be newly prepared substrates, or they may be substrates that have been previously processed in the cluster or in another device entirely. Similarly, depending on the processing required, the substrates 232 leaving the device 226 may be returned for subsequent patterning operations in the same lithography cluster, they may be sent to patterning operations in a different cluster, or they may be finished products to be sent for dicing and packaging.

[0051] Each layer of the product structure requires a different set of process steps, and the type of equipment 226 used at each layer can be completely different. In addition, even in the case where the processing steps to be applied by the equipment 226 are nominally the same, in a large facility, there may be several supposedly identical machines working in parallel to perform steps 226 on different substrates. Minor differences in settings or failures between these machines may mean that they affect different substrates in different ways. Even steps that are relatively common to each layer, such as etching (equipment 222), can be implemented by several etching equipment that are nominally the same but working in parallel to maximize yield. In addition, in practice, different layers require different etching processes, such as chemical etching, plasma etching, such as anisotropic etching, depending on the details and special requirements of the material to be etched.

[0052] Previous and / or subsequent processes may be performed in other lithography tools, as just mentioned, and may even be performed in different types of lithography tools. For example, some layers of a device manufacturing process that are very demanding on parameters such as resolution and overlay may be performed in more advanced lithography tools than other layers that are less demanding. Thus, some layers may be exposed in an immersion lithography tool, while others are exposed in a "dry" tool. Some layers may be exposed in a tool operating at a DUV wavelength, while others are exposed using EUV wavelength radiation.

[0053] In order to correctly and consistently expose the substrates exposed by the lithography apparatus, it is necessary to inspect the exposed substrates to measure properties such as overlay errors between subsequent layers, line thickness, critical dimensions (CD), etc. Therefore, the manufacturing facility where the lithography unit LC is located also includes a measurement system, which receives some or all substrates W that have been processed in the lithography unit. The measurement results are provided directly or indirectly to the supervisory control system SCS. If an error is detected, the exposure of subsequent substrates can be adjusted, especially when the measurement can be fast enough and rapid enough so that other substrates of the same batch are still to be exposed. In addition, the already exposed substrates can be stripped and reworked to improve the yield, or discarded to avoid further processing of the known defective substrates. In the case where only part of the target portion of the substrate is defective, additional exposure can be performed only on the good target portion.

[0054] Figure 1 Also shown is a metrology device 240, which is provided for measuring parameters of the product at desired stages of the manufacturing process. A common example of a metrology station in a modern lithography production facility is a scatterometer, such as a dark field scatterometer, an angle resolved scatterometer or a spectral scatterometer, and a scatterometer can be used to measure properties of the developed substrate at 220 before etching in the device 222. Using the metrology device 240, it can be determined that, for example, important performance parameters such as overlay or critical dimension (CD) do not meet the accuracy requirements specified in the developed resist. Prior to the etching step, there is an opportunity to strip the developed resist and reprocess the substrate 220 through the lithography cluster. By making minor adjustments over time by the supervisory control system SCS and / or control unit LACU 206, the measurement results 242 from the device 240 can be used to maintain accurate performance of the patterning operation in the lithography cluster, thereby minimizing the risk that the product does not meet specifications and requires rework.

[0055] Additionally, metrology equipment 240 and / or other metrology equipment (not shown) may be employed to measure properties of processed substrates 232, 234 and incoming substrate 230. Metrology equipment may be used on processed substrates to determine important parameters such as overlay or CD.

[0056] Typically, the patterning process in the lithography apparatus LA is one of the most critical steps in the process, requiring high-precision sizing and placement of structures on the substrate W. To ensure this high precision, the three systems can be combined into a so-called "holistic" control environment, such as Figure 2 As shown. One of these systems is the lithography apparatus LA, which is (in reality) connected to a metrology tool MET (a second system) and a computer system CL (a third system). The key to this "holistic" environment is to optimize the collaboration between these three systems to enhance the overall process window and to provide a tight control loop to ensure that the patterning performed by the lithography apparatus LA remains within the process window. The process window defines a range of process parameters (e.g., dose, focus, overlay) within which a particular manufacturing process produces a defined result (e.g., a functional semiconductor device) - typically the process parameters in the lithography process or patterning process are allowed to vary within these parameter ranges.

[0057] The computer system CL can use (a portion of) the design layout to be patterned to predict which resolution enhancement technology to use and perform computational lithography simulations and calculations to determine which mask layouts and lithography equipment settings achieve the maximum overall process window for the patterning process (in Figure 2 Typically, the resolution enhancement technique is arranged to match the patterning possibilities of the lithographic apparatus LA. The computer system CL may also be used to detect where within the process window the lithographic apparatus LA is currently operating (e.g. using input from a metrology tool MET) to predict whether defects may be present (e.g. due to suboptimal processing) Figure 2 ) is depicted in the second scale SC2 by an arrow pointing to “0”.

[0058] The metrology tool MET may provide input to the computer system CL to enable accurate simulations and predictions, and may provide feedback to the lithographic apparatus LA to identify possible drifts, e.g., in the calibration state of the lithographic apparatus LA (in Figure 2 ) is depicted by multiple arrows in a third scale SC3.

[0059] Various techniques can be used to improve the accuracy of copying patterns onto substrates. Accurately copying patterns onto substrates is not the only problem in IC production. Another problem is yield, which is generally a measure of how many functional devices a device manufacturer or device manufacturing process can produce per substrate. Various methods can be used to improve yield. One such method attempts to make the production of devices (e.g., using a lithography device such as a scanner to image a portion of a design layout onto a substrate) more tolerant to perturbations of at least one processing parameter during processing of the substrate, for example, in the process of using a lithography device to image a portion of the design layout onto a substrate. The concept of overlapping process windows (OPW) is a useful tool for this method. The production of devices (e.g., ICs) can include other steps, such as substrate measurement before, after, or during imaging, loading or unloading of substrates, loading or unloading of patterning devices, positioning the die under the projection optical device before exposure, stepping from one die to another die, and the like. In addition, various patterns on the patterning device can have different process windows (i.e., the process parameter space under which the pattern will be produced within the specification). Examples of pattern specifications associated with potential system defects include checking for necking, line pullback, line thinning, CD, edge placement, overlap, resist top loss, resist undercut and / or bridging. The process windows of all or some (usually patterns within a specific area) on a patterning device can be obtained by merging (e.g., overlapping) the process windows of each independent pattern. The process windows of these patterns are therefore called overlapping process windows. The boundaries of the OPW may contain the boundaries of the process windows of some independent patterns. In other words, these independent patterns limit the OPW. These independent patterns may be referred to as "hot spots", "critical features" or "process window limiting patterns (PWLP)", which are used interchangeably herein. When controlling the lithography process, it is possible to focus on hot spots, and it is usually economical. When there are no defects in the hot spots, it is likely that all patterns are defect-free. If the value of the processing parameter is outside the OPW, then when the value of the processing parameter is closer to the OPW, or if the value of the processing parameter is within the OPW, then when the value of the processing parameter is away from the boundary of the OPW, the imaging becomes more tolerant.

[0060] Figure 3 Exemplary sources of processing parameters 350 are shown. One source may be data 310 of a processing device, such as a source of a lithographic device, track, etc., parameters of a projection optics, substrate stage, etc. Another source may be data 320 from various substrate metrology tools, such as substrate height maps, focus maps, critical dimension uniformity (CDU) maps, etc. The data 320 may be acquired before applicable substrates are subjected to steps (e.g., development) for preventing substrate rework. Another source may be data 330 from one or more patterning device metrology tools, patterning device CDU maps, patterning device (e.g., mask) film stack parameter variations, etc. Yet another source may be data 340 from an operator of the processing device.

[0061] The control of the lithographic process is usually based on feedback or feedforward measurements, which are then modeled using, for example, inter-field (cross-substrate fingerprint) or intra-field (cross-field fingerprint) models. U.S. Patent Application 20180292761, incorporated herein by reference, describes a control method for controlling performance parameters such as overlay at the subfield level using an advanced correction model. Another control method using subfield control is described in European Patent Application EP3343294A1, also incorporated herein by reference.

[0062] However, while the advanced correction model may include, for example, 20 to 30 parameters, currently used lithographic apparatus (the term "scanner" will be used throughout the description for brevity) may not have actuators corresponding to one or more of these parameters. Therefore, only a subset of the entire parameter set of the model may be used at any given time. Furthermore, since advanced models require multiple measurements, it is undesirable to use these models in all cases, as the time required to perform the necessary measurements can reduce throughput.

[0063] Some major contributors to overlay error include, but are not limited to:

[0064] Scanner-specific errors: These may be caused by various subsystems of the scanner used during exposure of the substrate, in effect creating a scanner-specific fingerprint;

[0065] Process-induced wafer deformation: Various processes performed on the substrate may cause the substrate or wafer to deform;

[0066] Illumination setup differences: These are caused by the setup of the illumination system, such as the shape of the aperture, the position of the lens actuator, etc.

[0067] Heating effects: The heating-induced effects may differ between subfields of a substrate, especially for substrates where the subfields include different types of components or structures;

[0068] Reticle writing errors: errors may already exist in the patterning device due to manufacturing limitations; and

[0069] Topography Variations: The substrate may have topography (height) variations, especially around the edge of the wafer.

[0070] Instead of modeling the overlay error of the entire field, or as a complement to modeling the entire field, the overlay errors of independent subfields of the field can be modeled (e.g., at the die level or other functional area level). Although the latter requires more processing time, since both the field and the subfields therein are modeled, it allows correction of error sources related only to specific subfields as well as error sources related to the entire field. Other combinations (such as modeling the entire field and only certain subfields) are of course possible.

[0071] Even in cases where the errors are adequately modeled, actuation of synthetic corrections presents difficulties. Some corrections simply cannot be effectively actuated using the available control parameters (control knobs). Furthermore, while other corrections may be actuable, actually doing so may result in undesirable side effects. In essence, there are limits to what a scanner can actually do with a correction due to dynamic and control limitations and sensitivity.

[0072] Figure 4 A specific example of an inter-field overlay fingerprint that presents difficulties in actuation correction is shown. It shows a plot of the overlap OV (y-axis) versus the direction X (or Y). Each cross represents a measured overlay value, and each dot is the necessary corresponding compensating correction. The fitted line is the (near ideal) correction distribution, which fits the correction (dots). The sawtooth pattern shown in the overlay fingerprint is obvious; each part of the overlap varies essentially linearly, with X being a single die (the figure represents overlay measurements across 4 dies). The correction distribution follows (and therefore compensates for) the overlay fingerprint. This fingerprint is seen as a result of large stresses caused by large stacks, such as used in 3D-NAND or DRAM processes. This stress manifests itself both at the wafer level (leading to severe wafer warpage) and at the die level. At the die level, the overlay fingerprint includes the magnification inside each die. Due to the presence of multiple dies within one exposure field, the resultant field overlay fingerprint exhibits the sawtooth pattern shown (typically on the scale of tens of nanometers). Depending on the orientation of the device, the pattern can be a straight-through slit or a straight-through scan. Regardless of the orientation, the overlay cannot be corrected using the available models and actuators. In particular, it is not possible to actuate correction for such extreme patterns within the scanner alone.

[0073] Although embodiments herein will be specifically described in terms of overlay or edge position error (EPE), which manifests as a sawtooth pattern (e.g., caused by intra-die stress in 3D-NAND or DRAM processes, such as Figure 4 ), but it will be appreciated that it can be used to correct for any other higher order overlap, EPE or focal fingerprint.

[0074] To optimally correct Figure 4 It is important to be able to tune the scanner at spatial scales smaller than the pitch of the periodic distribution, e.g. Figure 4A "sawtooth" of a repetitive sawtooth distribution. Such independent sawtooth areas are typically associated with unit structures within independent tube cores. Therefore, the interface of the scanner should allow the definition of separately controllable areas within the exposure field. This concept is called a sub-field control interface; an example of which is disclosed in the above-mentioned European patent application EP3343294A1. For example, the control distribution of the wafer stage of a scanner configured for a first unit tube core / unit structure can be defined largely independently of the control distribution for a second unit / tube core structure further positioned along the scanning direction. The sub-field control infrastructure allows for more optimized correction of overlap (or focus) variations that are repeated at sub-field resolution. In addition, the ability to independently control different sub-field areas allows for mitigation of die-to-die or unit-to-unit variations in overlap / focus fingerprints within a tube core and / or within a unit.

[0075] Typically, scanner overlay control uses dynamic stage position control to adjust the placement of structures (features) such that overlay error is minimized. In principle, this can be achieved by pre-correcting for expected overlay error fingerprints (e.g., due to stress accumulation caused by applying subsequent layers) and / or by adjusting feature placement within subsequent layers to be adequately aligned with features in the previous layer(s).

[0076] An example technique for pre-correcting placement errors such as those caused by stress is to adjust the placement of features within the cell and / or the die on the mask. For example, this can be done during a mask manufacturing step and / or a mask correction step. More specifically, the control of the electron beam of the mask writing tool can be adjusted so that the features are offset relative to the nominal position. Ideally, the offset will be exactly opposite to the error offset being corrected, for example, a feature offset caused by deformation caused by stress after the subsequent layer is applied. The effect is that using such a mask will result in much fewer corrections to be made by the scanner overlay correction infrastructure. However, the corrections made by the mask are necessarily static and cannot account for any variations in the overlay fingerprint (e.g., field-to-field, wafer-to-wafer, and / or batch-to-batch variations). Such variations can be of the same order of magnitude as the fingerprint itself. In addition, there are actuation and sensitivity limitations in controlling such mask writing corrections that are inherent in the writing tool used (e.g., an electron beam tool or the like).

[0077] Scanner overlay correction is usually applied by the stage controller and / or the lens manipulator of the projection lens (odd aberration control can be used to control the placement of features). However, as already mentioned, the scanner cannot perfectly follow any desired overlay correction profile. One reason for this is that there are limits to the speed and acceleration that can be achieved by the wafer (and mask) stage. Another reason is the fact that the scanner exposes the substrate with a relatively large illumination spot (the so-called slit length, representing the spot size in the scanning direction, refer to EP application EP19150960.3, which is incorporated herein by reference). The extension of the spot means that during the scanning exposure, some parts of the features within the tube core / unit will always be in a suboptimal position, in which case the overlay correction is expected to be more than a simple movement across the entire tube core / unit. This change in the effective position (overlay) correction during the scanning operation effectively leads to a blurring of the spatial image of the feature, which in turn leads to a loss of contrast. This dynamic effect is usually called the moving standard deviation (MSD). The limitations on the positioning of the stage are usually related to the average position (overlay) error and are usually called the moving average (MA) error.

[0078] More specifically, the temporal moving average (MA) error and the temporal moving standard deviation (MSD) of the errors in the lithography phase are related to a critical time window, which includes the time interval during which each point on the die is exposed (in other words: receives a photon). If the average position error of a point on the die is high during this time interval (in other words: high MA error), the effect is a shift in the exposure image, resulting in an overlay error. If the standard deviation of the position error during this time interval is high (in other words: high MSD error), the image may show a tailing phenomenon, resulting in a fading error.

[0079] Both the average overlay error (MA) and the contrast loss due to MSD are contributors to the overall edge position error (EPE) budget and therefore require careful trade-offs when determining a certain control profile for the wafer and / or reticle stage; typically, a more MA-targeted control approach will result in a higher MSD impact, while an MSD-targeted control strategy may result in unacceptably large MA errors. EPE is a combined error caused by global critical dimension uniformity (CDU), local CDU (e.g., line edge roughness LER / line width roughness LWR), and overlay errors. It is these parameters that have the greatest impact on yield, as it is errors in these parameters that affect the relative positioning of features and whether any two features accidentally touch or fail to touch.

[0080] The present reticle correction is implemented without considering the scanner capabilities (or even without proper consideration of the sensitivity / capabilities of the reticle writing tool). This is followed by a subsequent exposure control optimization designed within the scanner, at which point the reticle correction is fixed. In the proposed method, reticle optimization and exposure control optimization are performed together to achieve improved performance based on performance metrics such as overlap, focus, CDU or EPE. Therefore, it is proposed to balance the control of the reticle formation process (e.g., writing, correction or modification process) with the control of the scanner exposure process (e.g., overlap control). In particular, a joint optimization of controls within these two control regimes is proposed.

[0081] For example, co-optimization of correction strategies for both the mask formation process and the scanner exposure process can determine an optimized mask correction that enables the co-optimized scanner correction to correct a simpler actuation overlay error distribution in the scan direction. Co-optimization can also take into account mask writing tool capabilities and / or sensitivities to better optimize mask correction. Furthermore, in one embodiment, it is proposed to co-optimize numerous components / processes of an IC manufacturing process and the interactions between them to obtain the best desired performance with respect to one or more performance parameters. Where multiple parameters are considered, they can be weighted by correlation.

[0082] In one embodiment, the reticle correction may be determined as a linear or step correction and the scanner exposure correction as a higher order (e.g., 3rd order or higher) correction. This enables the latter to be a smaller correction (e.g., lower in amplitude), thereby requiring less extreme motion of the stage(s), and thus easier to actuate within the scanner.

[0083] Figure 5 Such a method is conceptually illustrated. Figure 5 (a) depicts a desired correction 500 (dotted line) representing a "total correction" (initial spatial distribution) across subfields, e.g., the total correction distribution required to correct for an expected overlay residual of a die. According to the concepts disclosed herein, the desired correction 500 is decomposed into component spatial distributions, including a reticle subfield correction 510 component (linear component - dashed line) and a higher-order (but smaller amplitude) scanner subfield component 520 (solid line), which can be more optimally corrected by the scanner overlay correction infrastructure (e.g., stage and / or projection lens control).

[0084] When using linear reticle correction, it may not always be possible to correct for the total intra-die stress fingerprint, or it may be too easy to over-correct using linear correction. Figure 5(b) shows an alternative strategy for reticle subfield correction using step function correction. The desired correction 530 (dotted line) is shown, which is decomposed into a component spatial distribution, including a step function reticle subfield correction 540 component (dashed line) and a higher order scanner subfield component 550 (solid line). The step function helps to pull the edges of the higher order shapes (subfield components 550) to obtain better correction.

[0085] The reticle subfield correction can be implemented as a control recipe for a reticle writing tool, for a reticle correction tool, or other tools used in the manufacture, correction, or change of a reticle. In another embodiment, instead of (or in addition to) applying a permanent reticle writing modification (correction) during the reticle manufacturing step, a post-reticle manufacturing overlay correction can be applied to the reticle based on femtosecond laser pulse technology. Such a concept is described in U.S. Patent Application US20180307135, the entire contents of which are incorporated herein by reference.

[0086] More generally, the desired total correction distribution can be decomposed into various components (e.g., high-order, low-order, and scaled components), where these components are optimally assigned to scanner control or reticle manufacturing / correction control to achieve the best possible correction to minimize overlay residuals on the product.

[0087] Figure 6 is a flow chart describing an embodiment of the concepts disclosed herein. At step 610, a desired total correction distribution (initial spatial distribution) is obtained, which represents the distribution of overlay errors to be corrected along the scan direction. For example, this can be derived from an expected or known in-die stress fingerprint. At step 620, the total correction distribution is decomposed into a plurality of candidate scanner subfield correction distributions (candidate first component spatial distributions) and candidate reticle subfield correction distributions (candidate second component spatial distributions). At step 630, an optimal combination of the reticle subfield correction distribution and the scanner subfield correction distribution is selected based on a performance metric, for example, a combination that produces a minimum overlap residual on the field.

[0088] Figure 7 It shows the Figure 6A variation of the method of . Step 710 is the same as step 610. Step 720 may include designing a merit function that describes parameters in terms of one or more reticle subfield correction terms (e.g., a low-order or linear portion of a desired total correction) and one or more scanner subfield correction terms (e.g., a high-order portion of a desired total correction). In one embodiment, the scanner subfield correction terms may be constrained to a (e.g., relatively) small range to make them easier to actuate within the scanner, where the reticle subfield correction terms are allowed to have larger amplitudes. The smaller range in this context may include 50%, 40%, 30%, 20%, or 10% of the amplitude of the desired total correction distribution. Then, step 730 includes optimizing the merit function in terms of a performance parameter. For example, this may include solving an iterative algorithm that optimizes (e.g., minimizes) a performance parameter value (e.g., overlap or EPE).

[0089] The jointly optimized scanner correction profile may be further optimized within the scanner, for example based on scanner metrology (alignment / equalization etc.) or other metrology to determine the actual correction profile achieved during exposure.

[0090] The balance between reticle correction and scanner correction can be further refined by including optimization of the MA or MSD or MA / MSD combination associated with the scanner correction distribution with respect to a suitable performance parameter (e.g., expected EPE error of one or more critical features). In such an embodiment, the method can include identifying one or more critical features within a subfield and performing the optimization in terms of minimizing the expected EPE of the critical feature(s) and / or using the EPE of the critical feature(s) as an evaluation term / performance parameter in an evaluation function. Figure 6 or Figure 7 Any of the methods.

[0091] In other embodiments, the co-optimization step may consider (e.g., co-optimize) control parameters other than the stage / projection lens control profile in the scan direction, which affect the performance parameters. For example, the co-optimized scanner control profile may be further optimized in terms of focus, dose, etc., all of which also have an impact on EPE.

[0092] Instead of or in addition to dividing the correction profile into control profiles for different tools (scanner and reticle writing tool), the correction profile may also be distributed across multiple layers applied to the substrate.

[0093] In such an embodiment, the correction profile associated with a subfield of the exposure field can be decomposed into its fractional components for separate actuation in different layers on the substrate. Decomposing the correction profile in this manner and applying the fractional components in different layers enables determining the fractional components that are easier to actuate within the scanner and / or on the reticle. Thus, a proposed total correction profile to be implemented in the exposure of a first layer can be decomposed into complementary multiple control profiles for implementation in multiple layers. For example, a proposed total correction profile to be implemented in the exposure of a first layer can be decomposed into complementary first and second control profiles to be implemented in the exposure of the first layer and the second layer, respectively.

[0094] This embodiment can be implemented separately from the previously described embodiment such that the total correction profile is divided into at least a first correction profile and a second correction profile for the scanner. When exposing different layers on the substrate, the scanner will then implement the first correction profile and the second correction profile. In another embodiment, the first correction profile and the second correction profile can be configured as a control recipe (e.g., a reticle writing / correction tool) for reticle correction. In the latter case, the first correction is configured for a first reticle associated with a first layer and the second correction is configured for a second reticle associated with a second layer.

[0095] The concepts of assigning correction components to different layers and of reticle and scanner correction can be combined. This can be achieved by any suitable method. For example, reticle and scanner correction can be initially optimized jointly (e.g. using Figure 6 or Figure 7 ), and then further decompose one or both of the reticle and scanner corrections to be distributed over two or more layers. Alternatively, a layer distribution may first be performed based on the total correction to determine a layer correction distribution, and then each of the layer correction distributions may be decomposed into a reticle correction and a scanner correction. Alternatively, the co-optimization may simultaneously co-optimize the decomposition of the total correction distribution associated with a single layer into constituent reticle and scanner correction distributions for two or more layers (e.g., using a suitable merit function).

[0096] In all of the above examples, optimization can be performed in terms of any of the performance parameters (overlap, EPE, etc.) that have been disclosed. In one embodiment, the performance parameter error is associated with the stress caused by applying the subsequent layer, and the joint optimization is based on the expected performance parameter residual. The specification will continue to describe this embodiment in terms of the distribution of the control distribution between two layers. In all cases, the distribution of the control distribution can be between more than two layers, if applicable.

[0097] Figure 8Depicted is a decomposition of an overlay error distribution 800 across subfields into a first layer component 810 and a second layer component 820. An effective correction of such an overlay distribution may be achieved by defining a first layer overlay correction 830 for the first layer L1 and a second layer correction 840 for the second layer L2 following the first layer L1.

[0098] Typically, the total correction distribution is partitioned such that: a) the first correction and the second correction add up to a desired total overlap correction, and b) the first correction and the second correction can be implemented using a first control distribution and a second control distribution, respectively, which can be more accurately tracked using the control infrastructure of the scanner. For example, the first control distribution and the second control distribution can be constrained or kept within a range that is smaller than the total control distribution.

[0099] In one embodiment, the first layer control distribution and the second layer control distribution can be balanced so that a) the MSD characteristics associated with the first layer control distribution and the second layer control distribution do not excessively degrade the imaging quality of features within the first layer and the second layer, and b) the MA characteristics associated with the first layer control distribution and the second layer control distribution do not excessively degrade the overlap between the first layer and the second layer. Therefore, the overall desired overlap correction can be divided into multiple per-layer control distributions that are balanced to ensure that the per-layer EPE (MSD driven, indicating imaging quality) is acceptable and the overlap between layers remains within specifications (MA driven). This can be done based on overall acceptability (optimal ratio), or alternatively to obtain better correction in one layer (e.g., a critical layer) at the expense of another (e.g., less critical) layer.

[0100] This balance can be achieved by determining the maximum allowable MSD degradation and using it to optimize the scanner illumination settings. This can be done as part of a multi-focus imaging (MFI) mode of operation. The optimized scanner illumination settings can enable them to achieve the best in-die stress correction while maintaining an acceptable (e.g., practical) process window.

[0101] MFI uses multiple focus levels (e.g., via multiple wavelengths) to effectively increase the depth of focus (DoF) for a given numerical aperture (NA) of the objective lens. This can increase the imaging NA, thereby increasing the exposure latitude (process window), while the DoF can be optimized by the MFI based on the production layer requirements.

[0102] One problem with subfield control as described herein is that it tends to increase MSD. In this case, using MFI optimization can provide more room for MSD degradation, so any MSD constraints can be relaxed accordingly. For example, one method for decomposing the total control distribution can include performing any of the methods described above with respect to process parameters, particularly overlap (i.e., MA), where MSD is applied as a constraint (e.g., constrained to be within an allowed range). The MSD constraints can be relaxed by optimizing the irradiation via MFI to increase DoF and process window. Relaxing the MSD constraints increases the available correction space for the MA / overlap correction distribution, effectively allowing the platform to make more extreme movements.

[0103] Fig. 9 is a flow chart showing the proposed method for each layer. At step 900, the in-die stress performance parameter (e.g., overlap) fingerprint is measured and decomposed into layer scanner control components to be applied to two or more different layers (the method can also be applied to the scanner component of the first described embodiment for each tool to be decomposed). At step 910, the maximum allowed MSD is determined layer by layer based on the layer characteristics (typically, each layer has a different sensitivity to MSD), and based on this, the illumination conditions are optimized 920 to expose the layer. This can be achieved via controlling the scanner and / or laser (illumination source). At step 930, based on the decomposed layer control distribution and the optimized illumination conditions for each layer, an improved optimized scanner control distribution is determined for the layer. This can be fed back to the SIM in a simulation-based loop. Finally, at step 940, the performance on the actual product can be verified (e.g., via measurement) in an experimental-based loop and feedback EXP is given to it.

[0104] In this way, the correction range of the overlay (or other parameter of interest) sub-field correction can be extended by increasing the application process window, thus allowing a higher MSD limit.Such an optimization can be performed per layer.

[0105] Figure 8 and corresponding paragraphs in this document describe a method for splitting the correction distribution into multiple components; each component is associated with a specific layer, and the sum of the components adds up to the desired control distribution. A problem with this approach may be that the variability (instability) of the processing steps performed between the patterning of a first layer and the subsequent second layer may prevent the first component of the correction distribution associated with the first layer from being effectively (sufficiently accurately) determined. Any choice of decomposition made may result in undesirable performance degradation if the above-mentioned (intermediate) processing steps (e.g., deposition, etching, CMP, etc.) give rise to unexpectedly large or small in-plane distortion distributions (e.g., corresponding to sub-fields) across the tube core on the substrate.

[0106] To overcome this problem, it is proposed to use a double exposure technique, in which a first group of dies (corresponding to a group of subfields) is exposed in a first exposure, and a second group of dies (subfields) is exposed in a second exposure. Typically, the exposed dies are provided to a single layer on the substrate. Typically, each group of dies is associated with a different patterning device (e.g., mask, reticle). Fig.15a a, b, c depict a set of sub-fields and their corresponding die and their distribution across (2 in this case) reticles. Fig.15a An exposure field 1500 (dashed rectangle) is depicted covering an area of ​​the reticle in which 14 dies have been provided. The exposure field 1500 is divided into seven subfields 1501-1507. Fig.15b A first set of sub-fields {1511, 1512, 1515, 1517} configured to expose 4 rows of odd numbered die {1, 3, 5, 7} on a first reticle 1510 is depicted. Fig.15c A second set of subfields {1522, 1524, 1526} on a second reticle 1520 configured to expose uniformly numbered die {2, 4, 6} is depicted. By grouping the die into a first group including odd-numbered die and a second group of even-numbered die, and providing these groups to separate reticles, the distance between subfields (and dies) along the scan direction 1550 increases by at least the length of the subfield (the dimension along the scan direction 1550). The lithography tool may, for example, first expose one or more substrates after loading a first reticle and thus expose a first set of die to the substrate(s) and in a second exposure step (after loading a second reticle) expose a second set of die to the already exposed substrate(s).

[0107] The main advantage is that the lithography tool must apply (per exposure step) a significantly smaller spatially fluctuating control profile. This lower frequency control profile is better suited to the precise implementation of the stages and lens actuators available within the lithography tool. This Fig.16. A first (initial) control profile is associated with a performance parameter variation 1600 along the scan direction 1550 and across all subfields 1-7 (separated by dashed lines). The initial control profile (first control profile) must be able to follow the continuous line 1600 representing, for example, a typical sawtooth (overlapping) variation accurately enough. If the parameter variation 1600 is to be corrected / controlled in a single exposure, the lithography tool actuator needs to provide control (e.g., overlap) at a first spatial frequency "f1". If a double exposure approach is employed, a first control profile component 1615 (dashed line) may be applied to correct the performance parameter variation 1610 (solid line representing low pitch sawtooth) across a first set of subfields (1, 3, 5, 7), the first control profile (component) being applied during the first exposure step. Flat lines within the continuous line 1610 included between the triangular peaks indicate that essentially "don't care" or significantly reduced control requirements apply; the actuator may be operated in any manner to prepare for actuation during the next associated subfield. The actuator in the second exposure step can then be configured to correct or mitigate the performance parameter variation 1620 (solid line), corresponding to the second control component distribution 1625 (dashed line). The resulting control distribution (the sum of the first control distribution component and the second control distribution component) effectively completes the correction / control of the performance parameter variation 1600. The first control distribution component and the second control distribution component are then associated with a spatial pitch greater than the initial control distribution (e.g., a smaller spatial frequency "f2"). If the control distribution is divided into two components, the corrected spatial frequency "f2" is "f1" / 2 (half of "f1"). In the event that the spatial frequency required for control needs to be further reduced, it is recommended and in line with the spirit of the present invention to further divide into more than two control distribution components associated with more than two different sub-field groups. For example, a set of 4 reticles can be used, each reticle including a set of specific tube cores associated with a specific layout of the sub-field. This may be relevant if the number of tube cores along the scanning direction 1550 is very large (e.g., >>7).

[0108] It can be observed that the independent (per-exposure) control profile components 1615 and 1625 are selected so that they are continuous and, because they are each associated with a different set of subfields, their frequency is lower than would be necessary if the control profiles needed to provide correction of the performance parameter variation 1600 in only one exposure (e.g., for all subfields). The accomplished pitch relaxation and continuous behavior of the independent control profile components ensures better correctability of the performance parameter variation 1600 by the actuators (e.g., lenses and stage control infrastructure) of the lithographic apparatus (tool).

[0109] In another implementation, a group of dies is provided to one reticle, and the lithography tool exposes one reticle in a first exposure step, the reticle or wafer stage is shifted by an amount such that a group of dies is provided to the reticle in a second exposure step between the dies already provided. The advantage is that only one reticle is needed, but this is only feasible if the lithography tool is provided with a wafer (substrate) or reticle stage that allows a sufficiently large offset (e.g., at least corresponding to the length of a subfield). This is in Fig.17 The reticle 1700 is positioned at a first position 1710 relative to the substrate 1750 at the beginning of the first exposure step and at a second position 1720 at the beginning of the second exposure step. In this particular case, the difference 1720-1710 corresponds to a value that is the same or close to the dimension of the subfield 1511 along the scanning direction 1550.

[0110] In another implementation, only one mask (including Fig.15a all dies depicted in the image), including both odd and even dies, but instead of switching or moving the mask between exposures ( Fig.17 ), the lithography tool exposes the odd-numbered dies first, for example, by closing the mask (REMA) blade when the even-numbered dies are within the illumination beam, and exposes the even-numbered dies in a second exposure (again by using REMA to prevent exposure of unwanted dies (in this case, odd-numbered dies)).

[0111] To allow exposure of a subset of subfields of an exposure field to be performed, the lithography tool may be provided with a control interface configured to receive a layout of subfields to be exposed in one exposure step. The skipped subfields may then be associated with a turn-off REMA setting to prevent exposure of unselected subfields.

[0112] In one embodiment, a method for controlling a lithography process for exposing a pattern on an exposure field is provided, the method comprising: obtaining an initial spatial distribution associated with spatial variations of a performance parameter across the exposure field; decomposing the initial spatial distribution into at least a first component spatial distribution for controlling the lithography device to correct spatial variations of the performance parameter across a first plurality of subfields in a first exposure step and a second component spatial distribution for controlling the lithography device to correct spatial variations of the performance parameter across a second plurality of subfields in a second exposure step, wherein controlling the lithography process using the first component spatial distribution and the second component spatial distribution during corresponding exposure steps mitigates spatial variations of the performance parameter across the exposure field.

[0113] In one embodiment, the first plurality of sub-fields and the second plurality of sub-fields are associated with a die layout on a patterning device (reticle, mask).

[0114] In one embodiment, the first plurality of subfields is associated with a first patterning device and the second plurality of subfields is associated with a second patterning device.

[0115] In one embodiment, the first plurality of sub-fields and the second plurality of sub-fields are associated with a die layout on a patterning device (reticle, mask), wherein the second plurality of sub-fields are associated with a different position of the patterning device relative to the substrate compared to a position of the patterning device associated with the first plurality of sub-fields relative to the substrate used in a lithography process.

[0116] In an embodiment, the controlling of the lithography process is controlling a stage and / or a projection lens actuator used during the lithography process.

[0117] In one embodiment, a first plurality of subfields are associated with a first group of dies on a patterning device and a second plurality of subfields are associated with a second group of dies on the patterning device or another patterning device, wherein relative positions of the first group of dies with respect to the patterning device are staggered with respect to positions of the second group of dies with respect to the patterning device or another patterning device.

[0118] In one embodiment, the lithography tool is provided with a control interface configured to receive information associated with a definition of a subset of subfields comprised within an exposure field, the subsets being associated with subfields scheduled to be exposed in an exposure step by the lithography tool.

[0119] The present invention can be used to decompose an initial control profile (intended to correct for spatial variations of a performance parameter across multiple subfields within a field) into multiple components. In this case, the decomposition is based on separating the correction into a first exposure step and a second exposure step for patterning a layer associated with the performance parameter variation. The first exposure and the second exposure are associated with different groups of subfields, for example, distributed across multiple reticles. The control profile components can be further jointly optimized to provide a better correction for the observed spatial variations of the performance parameter across the subfields within the field of interest.

[0120] In one embodiment, a method for controlling a lithography process for exposing a pattern on an exposure field is provided, the method comprising: obtaining an initial spatial distribution associated with a spatial variation of a performance parameter of a first layer across subfields of the exposure field; decomposing the initial spatial distribution into at least a first component spatial distribution that can be used to control a lithography device to correct for the spatial variation of the performance parameter and a second component spatial distribution that can be used to control the lithography device to correct for the spatial variation of the performance parameter; and jointly optimizing the first component spatial distribution and the second component spatial distribution to achieve better correction of the spatial variation of the performance parameter compared to correction based on controlling the spatial variation of the performance parameter of the lithography device using the initial spatial distribution, wherein the first component spatial distribution is associated with a first exposure for patterning the first layer, and the second component spatial distribution is associated with a second exposure for patterning the first layer.

[0121] In one embodiment, the first component spatial distribution is associated with a first reticle associated with a first exposure, and the second spatial distribution is associated with a second reticle associated with a second exposure.

[0122] In one embodiment, the first exposure is associated with a first set of subfields or dies on a reticle and the second exposure is associated with a second set of subfields or dies on the reticle or another reticle.

[0123] In one embodiment, the first reticle and the second reticle include a first die region associated with a sub-field having no control requirements applied or having reduced control requirements applied and a second die region having control requirements applied.

[0124] In one embodiment, a first die region on a first reticle is associated with a subfield on a second reticle that is different from the first die region.

[0125] In summary, a method is proposed that divides a correction profile into multiple component correction profiles based on co-optimization between tools, processes and / or layers that form part of an IC manufacturing process on a substrate. This co-optimization will improve the ability to achieve (actuate) corrections through co-optimized control of tools, processes and / or layers. For example, the division of corrections can be between processes such as a mask writing process and a scanner control process.

[0126] It can be shown that such an approach will improve overlay and therefore yield. Furthermore, optimized reticle correction can reduce the need for expensive updates of the reticle (less reticle correction is required due to the flexibility of the optimized scanner control). The extension of the overlap sub-field correction range increases the flexibility of the fingerprint within the reticle die, which in turn further increases the potential for overlay correction. The reticle write fingerprint can be jointly optimized, thereby benefiting all key overlay optimizer parameters, thereby improving product performance in one or more of EPE, imaging, CDU, overlay, focus and leveling. The scanner workload can also be split so that only a small part of the correction range needs to be addressed in any single layer, while the remaining part(s) are addressed in one or more subsequent layers.

[0127] As mentioned above, and in Figure 4 As depicted in Figure 1, the overlay of each subfield / die exhibits a sawtooth pattern that requires a corresponding aliasing correction. The required overlay correction is highly discontinuous between dies or subfields, which is very challenging in scanner overlay control. Essentially, there is only a small amount of time and space during exposure to achieve such a large overlay error correction (e.g., including sign changes).

[0128] A further proposal to solve this problem will now be described. In this embodiment, it is proposed to determine an optimized reticle correction per subfield, which is optimized over the field (e.g., in at least one direction) taking into account the scanner actuation capabilities. The optimization may be such that, after writing the reticle, the remaining overlay error distribution to be corrected is better (e.g., easier) to correct by the scanner control infrastructure than with the current reticle correction strategy.

[0129] Fig.10 This embodiment is conceptually shown. Fig.10 (a) shows a typical current strategy. The solid black lines 1000a, 1000b each show the reticle correction distribution for two adjacent sub-fields or dies (generally similar in this example) of a field. The dashed / grey line 1010 shows the required scanner correction distribution produced over the two sub-fields (e.g., for a field exhibiting a sawtooth pattern or the like). It can be seen that this correction is discontinuous between sub-fields and therefore may not be correctable by an essentially continuous control system (such as stage control during scanning operation).

[0130] Fig.10 (b) shows an improved reticle correction profile over two adjacent fields. Such a reticle correction profile may be determined by jointly optimizing the reticle correction profile of each subfield over two or more subfields of a field; for example, for all subfields of a field in one or two directions parallel to the substrate / reticle plane. Fig.10In (b), the reticle correction profiles 1020a, 1020b are such that it is now easier to actuate the required scanner correction profile 1030. For example, the second reticle distribution 1020b causes (e.g., overlay) errors in its corresponding subfield, which the scanner can more easily correct from its instantaneous state / configuration after correcting the previous subfield, where the reticle correction profile 1020a has been applied.

[0131] In a particular example such as shown here, optimization of two adjacent subfields can result in at least one overcompensated reticle correction distribution 1020b that is substantially inverted compared to the nominal correction, such that the resulting residual overlay error distribution tends to exhibit a smooth or smoother curve over the two subfields, which will actuate better within the scanner. Of course, this concept can be extended to more than two subfields.

[0132] While clearly relevant in terms of subfields / die along the scan (Y) direction, it will also be appreciated that this embodiment can also be applied along the slit (X) direction. Scanner correction capabilities in this direction are more limited (e.g., limited to lens correction, etc.), so there may be additional benefits to performing this optimization in the X direction.

[0133] As already described, the reticle correction profile can be applied by determining a specific reticle writing correction for a reticle writing tool.

[0134] This embodiment can be combined with any of the above embodiments. For example, the mask correction co-optimization over multiple subfields can also be co-optimized with the scanner control distribution optimization (e.g., using Figure 5 This may, for example, result in adjacent linear mask correction profiles having slopes opposite to each other (corresponding to Figure 5 (a)) or the reverse step size (corresponding to Figure 5 (b)). In addition, the joint optimization of mask corrections over multiple subfields can also be decomposed over multiple layers (e.g., using Figure 8 described method).

[0135] In one embodiment, a method for determining an overlay error correction is disclosed, the method comprising: obtaining an overlay error distribution across a plurality of subfields; determining a reticle correction across at least two adjacent subfields of the plurality of subfields, wherein the reticle correction is based on an expected improvement in continuity of the overlay error distribution between the adjacent subfields. In one embodiment, the reticle correction is configured to achieve continuity of the overlay error distribution across the adjacent subfields by substantially inverting and / or overcorrecting the overlay error distribution.

[0136] Typically, after processing, the substrate (wafer) exhibits a certain global overlapping fingerprint across the wafer surface. This global fingerprint is usually most obvious in the peripheral or edge area of ​​the wafer. The global fingerprint is modeled by a suitable model and further controlled using "advanced process control" (also known as APC) control methods, which are usually run-to-run control solutions, where the global overlapping fingerprints observed in the first batch of wafers are used to correct the processing (exposure) of subsequent batches of substrates. More information about APC can be found in U.S. Patent Application Publication No. US20120008127, which is incorporated herein by reference. A typical APC implementation involves regular measurements of the metrology features on the substrate to monitor and correct drift associated with one or more processing equipment. In run-to-run (usually abbreviated as run2run) APC control, fingerprints (e.g., overlapping fingerprints) are estimated based on a group of substrates (e.g., wafers) measured in each batch. One or more measurement fields from these substrates are suitable for fingerprints, and then the fingerprints are usually mixed with earlier fingerprints to create new fingerprint estimates using an exponentially weighted moving average (EWMA) filter. Alternatively, the fingerprint can simply be updated regularly, or even measured once and kept constant. Combinations of some or all of these methods are also possible. The results of this calculation are then run through an optimization job to set one or more scanner actuators and / or other tool actuators / settings for the next batch to reduce or minimize overlap.

[0137] In addition to the global APC overlay correction, subfield specific corrections can be defined based on higher density overlay data, as already described. For example, Figure 4 The sawtooth overlay error distribution shown may require individual correction for each subfield. This may involve one or more of: scanner correction at the subfield level, reticle correction at the subfield level, distribution of corrections for each subfield across multiple layers of the process. Thus, there may be at least two active control infrastructures for overlay control on the same set of wafers (batch). A typical case where there is a risk of interference of the two control mechanisms is the control of edge dies, which are subject to large edge effects associated with global fingerprints and significant intra-die effects associated with intra-die stresses.

[0138] Preferably, the two control systems (global or APC control loop and subfield-based control loop) do not interfere to the extent that overcorrection or undercorrection of overlay errors occurs (e.g., the two control loops compete with each other). To avoid this, it is proposed in a further embodiment that the global correction and the subfield correction are determined separately in a joint optimization; for example, it is realized that either the global correction or the subfield correction is applied. This approach can be applied to the entire substrate, or optionally only to certain areas, such as edge fields or edge tube cores.

[0139] Fig.111 is a flow chart describing the principles behind the control method according to an embodiment. Global (e.g., APC) measurement and / or correction data 1100 and sub-field (e.g., in-die) measurement and / or correction data 1110 are received. The in-die data is decomposed 1115 into sub-field component data 1120 and global component data 1130. The sub-field component data 1120 can describe the data in terms of changes in the entire component or only with respect to the global component data 1130. The sub-field component data 1120 and the global component data 1130 are used in a co-optimization step 1140, which determines a preferred control strategy. This step may include determining appropriate weights of global (e.g., APC) corrections relative to sub-field control; for example, such that overlap errors (residuals) after co-optimization corrections are minimized and / or in terms of optimization strategies for dynamic and static corrections. As a result, the co-optimization step 1140 results in sub-field corrections 1150 and global corrections, the latter of which are subtracted from the global correction data 1100 (so that they are not actuated twice) to obtain a global (run-to-run) correction 1160 (eg, as part of an APC control loop).

[0140] Fig.12 (a) conceptually illustrates the decomposition step 1115 according to an embodiment. The fingerprint 1200 across the wafer is decomposed into a global fingerprint 1210 (eg, which may be corrected by an APC controller) and a sub-field component 1220 (eg, which may be corrected by a sub-field control infrastructure).

[0141] Fig.12 (b) Conceptually illustrates an example overlap correction strategy based on fitting a global model before fitting residuals on a sub-field basis. A global wafer fingerprint 1260 is obtained by extracting the intra-field (including intra-die) component 1250 from the cross-wafer fingerprint 1240. This clean global wafer fingerprint 1260 is used as input to the APC control infrastructure, while the intra-field content 1250 is used for the sub-field control infrastructure. Alternatively, a portion of the edge fingerprint of the global overlap error fingerprint can be assigned to the sub-field control infrastructure, which results in less overlap error content that the APC control infrastructure needs to correct. This is in Fig.12 This is shown in (c) where the sub-field control distribution 1270 shows a larger variation (across the wafer) due to the edge effects now included therein, which results in a smaller variation in the global component 1280.

[0142] Thus, a method for determining a jointly optimized global and subfield control strategy for a lithography process is disclosed, the method comprising a) acquiring global measurement data and intra-die measurement data, b) decomposing the intra-die measurement data into global component data and subfield component data, c) determining a jointly optimized control strategy using the global component data, the subfield component data, and known scanner control characteristics, and d) providing a global control input and a subfield control input based on the jointly optimized control strategy. The determination of the jointly optimized control strategy may include determining appropriate weightings of the global component data and the subfield component data or corrections thereof.

[0143] like Figure 4 As shown in FIG. 1 , the stress within the die usually causes a significant sawtooth distribution of overlay error along the scanning direction. The control profile of the scanner used to correct this overlay distribution (repeated across multiple dies in an exposure field) also reflects this typical sawtooth shape. Fig.13 , illustrated by the line showing the large amplitude control distribution (usually a periodic synchronization signal is applied between the mask and the wafer stage). The remaining overlay error (MA = moving average) is shown by the small amplitude dashed line graph (amplitude is about 2 nm). The dynamic error (MSD = moving standard deviation) in the scanning direction (in this case the Y direction) is shown by the small amplitude top (thick line) graph. MSD is also often called "fading" and can basically be seen as high frequency vibrations between the mask and the wafer stage, which have the effect of smearing the aerial image of the projected pattern and causing a loss of contrast. This effect of reduced contrast in the aerial image is usually related to a dimensional change of the feature (pattern) of interest. This dimensional change is direction specific (e.g. MSD-Y in the scanning direction mainly causes dimensional changes in the Y direction) and, in general, the dimensional change is proportional to the amplitude of the MSD. Therefore, MSD-Y changes across a field or tube core will cause dimensional changes across the tube core, such as dimensional changes of contact holes. In the case of two-dimensional structures (such as contact holes), this manifests itself as field coordinate related aspect ratios of the X and Y dimensions of the two-dimensional structure. In the case of contact holes, the ratio between the X and Y dimensions is often referred to as the ellipticity of the contact hole. Fig.13 The MSD-Y variation shown, due to the strong within-die overlay correction (control distribution) will lead to within-die (and of course within-field) contact hole ellipticity variation.

[0144] Fig.14A method suitable for at least partially correcting deformation variations of contact holes within a tube core according to the present invention is shown. First, the MSD-Y distribution is considered based on the analysis of the required control distribution (to correct the overlay error distribution). Secondly, the effect of MSD-Y variations on the contact hole ellipticity variations is calculated (for example, using a lithography simulation tool) or measured. Third, the MSD-X distribution is designed so that the expected contact hole ellipticity is restored to the nominal value. Basically, an MSD is introduced, which has at least a component perpendicular to the scanning direction, thereby affecting the size of the contact hole in the X direction. For example, this can be achieved by introducing high-frequency oscillations in the X direction between the mask and the wafer stage, the amplitude of which depends on the field coordinate (for example, adjusting the amplitude so that the MSD-X level is equal to the scanning-induced MSD-Y level). Alternatively, the projection lens aberration can be adjusted by configuring the lens manipulator to induce a field-dependent MSD-X component, for example by causing a field coordinate-dependent pattern shift in the x-direction. As a result, the contact hole is no longer elliptical. However, due to the presence of MSD-X and Y, the radius of the contact hole may deviate from the nominal value, resulting in a change in the radius of the contact hole along the scanning direction. If this variation exceeds a certain tolerance, a fourth step may be required: applying field coordinate-dependent dose correction during the scanning operation to maintain the size of the contact hole at the target. Alternatively, the scanning speed can be changed to adjust the cumulative dose at each field point, or the z position (position) can be dynamically adjusted to maintain the contact hole radius at the target.

[0145] Instead of using the lithography equipment control infrastructure (lens and / or platform synchronization) to compensate for pattern deformation variations along the scan direction, the pattern device (reticle) can also be configured to pre-correct the effects of decay on the fidelity of the pattern geometry. For example, a reticle can include multiple tube cores oriented along the scan direction and including contact hole features. By pre-compensating the ellipticity variations of the contact holes on the reticle, contact holes of nominal shape can be obtained after exposure of the contact holes in the resist. It is then necessary to manufacture the reticle for all tube cores of interest based on the knowledge of the ellipticity variations (fingerprints) caused by decay along the scan direction. This can be achieved by changing the shape of the features (contact holes) on the reticle or by OPC methods (e.g., placing auxiliary features in one or more directions). In addition, the size of the contact hole can be selected to further control the radius of the contact hole after exposing the contact hole to the photosensitive layer (resist) on the substrate.

[0146] Control of the MSD-X is typically part of controlling the sub-field type of the lithographic apparatus, as the MSD-X(t) control signal may be die specific.

[0147] In addition to control aspects, metrology plans can also be generated based on known dynamic characteristics; for example based on the known change in feature size due to scan induced fading (MSD-Y). Fig.13It is shown that the MSD-Y level is highly dependent on the position within the die. Therefore, any control of fading-induced dimensional variations will benefit from metrology data associated with positions along the scan direction where the MSD-Y amplitude is expected to be significant. In one embodiment, it is recommended to provide CD (critical dimension) measurements at one or more locations where the scan-induced MSD results in a dimensional deviation greater than a threshold. The threshold can be associated with the accuracy of the metrology tool that determines the CD value, for example the threshold can be 0.2nm.

[0148] In one embodiment, a method of configuring a process associated with fabricating a semiconductor device on a substrate is disclosed, the method comprising: determining an expected variation in a ratio between dimensions of features provided to the substrate using a patterning process; and configuring the patterning process and / or a patterning device used in the patterning process to at least partially correct for the expected variation in the ratio across regions on the substrate.

[0149] The expected changes may be due to asymmetric fading changes, such as having MSD changes in only one direction (due to asymmetry in platform dynamics and / or lens aberration changes).

[0150] The ratio between the dimensions may be, for example, the ellipticity of the contact hole or the aspect ratio of a geometry such as a line or space.

[0151] Configuration of the patterning process may include adjusting the stage synchronization (between the mask stage “RS” and the wafer stage “′WS”) so that MSDx(t) is coupled to MSDy(t) and the vector [MSD-X(t), MSD-Y(t)] is always at the same angle to the X-axis.

[0152] Configuration of the patterning process may also involve applying a wavefront variation during scanning such that the introduced wavefront asymmetry WA(t) counteracts the effect of MSDx / y(t) on the ratio between the dimensions.

[0153] Configuration of the patterning process may also include dose and / or scan speed control with the goal of bringing feature size back to target (eg, avoiding CD variations).

[0154] Configuring the patterning process may also involve platform control aspects, such as: providing a platform control distribution (x(t), y(t), z(t)) that meets the following criteria: a) ratio variations (ellipticity contact holes) should be minimized, and b) CD remains on target (scanning speed is slower when more dose is needed to bring the size back to target), and c) configuring z(t) to increase the MSD-Z (contrast) term, where no (or very little) MSD-X / Y is introduced to obtain constant contrast across the tube core and / or exposure field, and constant size of the exposed features.

[0155] The configuration of the patterning device can involve a field coordinate-dependent ratio of feature sizes that offsets expected variations in this ratio due to patterning process limitations.

[0156] The configuration of the patterning device may involve field coordinate dependent OPC configured to selectively deform features such that there is no change in the ratio between dimensions after the patterning process.

[0157] Configuration of the patterning apparatus may involve field coordinate dependent reticle transport to keep the CD presented to the features of the wafer constant (except for MSD-X / Y balance; hence the symmetry imposed by scanner control and CD offsets corrected by reticle transport tuning).

[0158] Alternatively, for each die (subfield), a control profile can be derived that provides a better balance between fade induced fading (MSD-Y) CD variation and overlap variation. Then, for each subfield, the control profile is optimized as Fig.13 The control distribution shown is configured to provide the smallest possible EPE, which is an aggregate metric that includes both overlap and CD variation effects. It is recommended to configure the control distribution so that the composite CD variation due to the fading component and the composite overlap residual (MA component) provides the minimum EPE error for the feature of interest.

[0159] In another embodiment, a method of determining a sampling scheme associated with an exposure field on a substrate undergoing a patterning process is disclosed, the method comprising: deriving an expected distribution of pattern size errors and / or pattern positioning errors across the exposure field due to a time dependency of synchronization errors in movement of a patterning device and a substrate during the patterning process; and determining the sampling scheme based on an expected accuracy of characterizing the expected distribution using measurements of pattern size and / or pattern position according to the sampling scheme.

[0160] Further embodiments of the invention are disclosed in the following list of numbered clauses:

[0161] 1. A method for determining a spatial distribution of a component for controlling a lithography process for exposing a pattern on an exposure field, the method comprising:

[0162] obtaining an initial spatial distribution describing a spatial variation of a performance parameter of a first layer across at least a subfield of the exposure field; and

[0163] decomposing the initial spatial distribution into component spatial distributions associated with the subfields, the component spatial distributions comprising at least a first component spatial distribution and a second component spatial distribution; and

[0164] jointly optimizing the component spatial distribution to minimize the error of the performance parameter;

[0165] wherein at least the first component spatial distribution is usable by a lithographic apparatus to control a lithographic process for exposing a pattern on the exposure field in the first layer; and

[0166] The first component spatial distribution can be more accurately actuated by the lithographic apparatus than the initial spatial distribution.

[0167] 2. A method according to clause 1, wherein the performance parameter comprises or relates to an overlap of the first layer with respect to one or more other layers.

[0168] 3. A method according to clause 1 or 2, wherein the performance parameter comprises an edge position error.

[0169] 4. A method according to any preceding clause, wherein the first component spatial distribution is restricted to a range of values ​​that is smaller than the total range of the initial spatial distribution.

[0170] 5. A method according to any preceding clause, wherein the first component spatial distribution is optimized to be actuatable by the lithographic apparatus.

[0171] 6. A method according to any preceding clause, wherein the first component spatial distribution is used to control one or both of stage positioning control within the lithographic apparatus or projection lens control within the lithographic apparatus.

[0172] 7. A method according to any preceding clause, wherein the second component spatial distribution relates to a reticle correction of a reticle used in the lithography process to expose the pattern in the first layer.

[0173] 8. The method of clause 7, wherein the second component spatial distribution is used to control the manufacturing and / or modification of the reticle.

[0174] 9. A method according to clause 7 or 8, wherein the second component spatial distribution is used to control the relative positioning of features on the reticle.

[0175] 10. A method according to any of clauses 7 to 9, wherein the second component spatial distribution is of lower order than the first component spatial distribution.

[0176] 11. A method according to any one of clauses 7 to 10, wherein the second component spatial distribution is a linear or step distribution.

[0177] 12. A method according to any one of clauses 7 to 11, wherein the decomposition step further comprises decomposing the initial spatial distribution into layer component spatial distributions so as to distribute control based on the initial spatial distribution over a plurality of layers, the plurality of layers including the first layer.

[0178] 13. The method according to any one of clauses 7 to 11 comprises decomposing at least one of the component spatial distributions into layer component spatial distributions, each layer component spatial distribution being used to control the lithography process of a specific layer among a plurality of layers, the plurality of layers including the first layer, so as to distribute the control based on the at least one of the component spatial distributions over a plurality of layers.

[0179] 14. The method of clause 11, wherein the at least one of the component spatial distributions decomposed into layer component spatial distributions comprises a first component spatial distribution.

[0180] 15. A method according to any one of clauses 1 to 6, wherein the decomposing step comprises decomposing the initial spatial distribution into the component spatial distributions so as to distribute the control based on the initial spatial distribution over a plurality of layers, the plurality of layers including the first layer, and

[0181] wherein the second component spatial distribution is usable by a lithographic apparatus to control a lithographic process for exposing a pattern on the exposure field in a second layer, and wherein the second component spatial distribution is more accurately actuatable by the lithographic apparatus than the initial spatial distribution.

[0182] 16. A method according to any one of clauses 7 to 15, comprising using multi-focus imaging when performing the lithography process; and

[0183] Based on a maximum allowed moving standard deviation associated with said initial spatial distribution and / or an appropriate component spatial distribution, illumination settings are optimized during said exposure of a pattern.

[0184] 17. The method according to clause 16, wherein for each layer the maximum allowed movement standard deviation is determined and the illumination settings are optimized.

[0185] 18. A method according to any preceding clause, comprising designing an evaluation function that describes the performance parameter at least in terms of one or more first correction terms related to the first component spatial distribution and one or more second correction terms related to the second component spatial distribution; and

[0186] The merit function is iteratively solved to determine the first component spatial distribution and the second component spatial distribution.

[0187] 20. A method according to any one of clauses 1 to 17, wherein the initial spatial distribution is decomposed into a plurality of candidate first component spatial distributions and candidate second component spatial distributions; and the method comprises: determining an optimal combination of the first component spatial distribution and the candidate second component spatial distribution based on the performance metric.

[0188] 21. A method according to any preceding clause, wherein the joint optimization step comprises optimization of a moving average, a moving standard deviation or a combination of a moving average and a moving standard deviation associated with the initial spatial distribution.

[0189] 22. The method of clause 21, wherein the optimization of a moving average, a moving standard deviation, or a combination of a moving average and a moving standard deviation is determined with respect to minimizing an error of one or more features identified as having a smaller process window.

[0190] 23. A method according to any preceding clause, wherein the initial spatial distribution is derived from a known in-die stress fingerprint.

[0191] 24. A method according to any preceding clause, wherein the first component spatial distribution is third order or higher.

[0192] 25. A method according to any preceding clause, comprising performing the lithography process based on the component spatial distribution.

[0193] 26. A method according to any preceding clause, wherein each sub-field relates to a single die or a functional area of ​​said single die.

[0194] 27. A method according to any preceding clause, wherein the method is performed to determine a respective set of said component spatial distributions for each sub-field of said exposure field.

[0195] 28. A computer program comprising program instructions operable, when run on a suitable device, to perform the method according to any one of clauses 1 to 27.

[0196] 29. A non-transitory computer program carrier comprising a computer program according to clause 28.

[0197] 30. A lithographic apparatus operable to perform the method according to any of clauses 1 to 27.

[0198] 31. A method for determining an overlay correction distribution associated with a subfield within an exposure field, the method comprising:

[0199] obtaining an overlay error distribution between a first layer and a second layer across at least the subfield; and

[0200] A first overlay correction profile for use in manufacturing or modifying a reticle used in patterning the first layer and a second overlay correction profile for use in patterning the first layer are determined.

[0201] 32. A method for determining an overlap correction distribution, the method comprising:

[0202] obtaining an overlay error distribution between the first layer and the second layer across subfields within the exposure field; and

[0203] A first overlay correction distribution for the first layer and a second overlay correction distribution for the second layer are determined, wherein the first correction distribution and the second correction distribution are added to mitigate the overlay error distribution across the subfield and are jointly optimized to achieve desired imaging quality across both the first layer and the second layer.

[0204] 33. A method for determining an overlay error correction, the method comprising:

[0205] obtaining an overlay error distribution extending across a plurality of subfields along a scanning direction; and

[0206] A reticle correction is determined across at least two adjacent subfields of the plurality of subfields, wherein the reticle correction is based on an expected improvement in continuity of the overlay error distribution between the adjacent subfields.

[0207] 34. A method for providing an APC control input to an APC controller and a subfield control input to a lithographic apparatus, the method comprising:

[0208] a) obtaining a fingerprint of the parameters measured across the substrate;

[0209] b) decomposing the fingerprint into a global component and a subfield component;

[0210] c) determining a jointly optimized control strategy using the global component and the sub-field components and a control characteristic of the lithographic apparatus; and

[0211] d) Determining the APC control input and the sub-field control input based on the common optimization control strategy.

[0212] 35. A method for controlling a photolithography process for exposing a pattern on an exposure field, the method comprising:

[0213] obtaining an initial spatial distribution associated with a spatial variation of a performance parameter of a first layer across a subfield of the exposure field;

[0214] decomposing the initial spatial distribution into at least a first component spatial distribution that can be used to control a lithographic apparatus to correct for the spatial variation of the performance parameter and a second component spatial distribution that can be used to control the lithographic apparatus or another apparatus to correct for the spatial variation of the performance parameter; and

[0215] The first component spatial distribution and the second component spatial distribution are jointly optimized to achieve better correction for the spatial variation of the performance parameter compared to correction based on controlling the spatial variation of the performance parameter of the lithographic apparatus using the initial spatial distribution.

[0216] 36. A method for determining an overlay error correction, the method comprising:

[0217] obtaining an overlay error distribution extending across a plurality of subfields; and

[0218] Determining a reticle correction across at least two adjacent subfields of the plurality of subfields; characterized in that:

[0219] The reticle correction is based on an expected improvement in continuity of the overlay error distribution between the adjacent subfields.

[0220] 37. The method of clause 36, wherein the reticle correction is configured to achieve continuity of the overlay error distribution across the adjacent subfields by substantial inversion and / or overcorrection of at least part of the overlay error distribution.

[0221] 38. A method according to clause 36 or 37, wherein the reticle correction is determined for at least two adjacent fields in the scan direction and / or the slit direction.

[0222] 39. A method for determining co-optimized global and sub-field control strategies for a lithography process; comprising:

[0223] a) Obtaining global measurement data and intra-die measurement data;

[0224] b) decomposing the intra-die measurement data into global component data and sub-field component data;

[0225] c) determining the jointly optimized control strategy using the global component data, the sub-field component data and known scanner control characteristics; and

[0226] d) providing a global control input and a sub-field control input based on the common optimization control strategy.

[0227] 40. A method according to clause 39, wherein the sub-field component data is described according to variations of the global component data to which it corresponds.

[0228] 41. A method according to clause 39 or 40, wherein step c) comprises determining appropriate relative weightings of, and / or corrections thereto, the global component data and sub-field component data.

[0229] 42. A method according to clause 39, 40 or 41, wherein at least part of the edge fingerprint of the global measurement data is assigned to be corrected via the sub-field control input.

[0230] 43. A method of configuring a process associated with fabricating a semiconductor device on a substrate, the method comprising:

[0231] determining an expected change in ratios between dimensions of features provided to the substrate using a patterning process; and

[0232] The patterning process and / or the patterning device used in the patterning process are configured to at least partially correct the expected variation in the ratio across areas on the substrate, wherein the configuration includes one or more of: a balance in position control of the substrate along a scanning direction and a non-scanning direction, a balance in aberrations of a projection lens used in the patterning process, and a balance in dimensions of structures along a first direction and a second direction.

[0233] 44. A method of determining a sampling scheme associated with an exposure field on a substrate undergoing a patterning process, the method comprising:

[0234] deriving an expected distribution of pattern size errors and / or pattern positioning errors across the exposure field due to a time dependency of synchronization errors between movement of a patterning device and the substrate during the patterning process; and

[0235] The sampling scheme is determined based on a desired accuracy of characterizing the expected distribution using measurements of the pattern size and / or the pattern position according to the sampling scheme.

[0236] 45. A method for controlling a lithographic apparatus, the lithographic apparatus being configured to pattern an exposure field on a substrate comprising at least a sub-field, the method comprising:

[0237] obtaining an initial spatial distribution associated with a spatial variation of a performance parameter associated with a first layer on the substrate across at least a subfield of the exposure field; and

[0238] The initial spatial distribution is decomposed into at least a first component spatial distribution for controlling the lithographic device at a first spatial scale and a second component spatial distribution for controlling the lithographic device at a second spatial scale associated with the size of the sub-field, wherein the decomposition includes: jointly optimizing the first component spatial distribution and the second component spatial distribution based on correcting the spatial variation of the performance parameter across the sub-field.

[0239] 46. ​​The method of clause 45, wherein the joint optimization is further based on an imaging quality metric.

[0240] 47. A method according to clause 45 or 46, wherein:

[0241] The step of acquiring the initial spatial distribution includes: acquiring in-die measurement data and global measurement data, wherein the in-die measurement data at least includes the initial spatial distribution;

[0242] The decomposing step comprises: decomposing the in-die measurement data into the first component spatial distribution including a global component and the second component spatial distribution including a sub-field component; and

[0243] The joint optimization step comprises using the global component, the sub-field components and known lithographic apparatus control characteristics.

[0244] 48. The method of clause 47, further comprising providing a global control input and a subfield control input to the lithographic apparatus based on the jointly optimized global and subfield components.

[0245] 49. A method according to any of clauses 45 to 48, wherein the performance parameter comprises or relates to an overlap of the first layer relative to one or more other layers.

[0246] 50 A method according to any one of clauses 45 to 49, wherein the performance parameter comprises an edge position error.

[0247] 51. A method according to any of clauses 45 to 50, wherein the first component spatial distribution is restricted to a range of values ​​that is smaller than the range of values ​​included in the initial spatial distribution.

[0248] 52. A method according to any of clauses 45 to 51, wherein the joint optimization is further based on bringing the first component spatial distribution within an actuatable range of the lithographic apparatus.

[0249] 53. The method according to any one of clauses 45 to 52 further includes obtaining an evaluation function, which describes the correction of the performance parameter at least in terms of one or more first correction terms related to the first component spatial distribution and one or more second correction terms related to the second component spatial distribution, and wherein the joint optimization is based on iteratively solving the evaluation function to determine the first component spatial distribution and the second component spatial distribution.

[0250] 54. A method according to any of clauses 45 to 53, wherein the joint optimization is further based on optimization of a moving average, a moving standard deviation or a combination of a moving average and a moving standard deviation associated with the initial spatial distribution.

[0251] 55. A method according to any of clauses 45 to 54, wherein the initial spatial distribution is derived from a known in-die stress fingerprint.

[0252] 56. A method according to any of clauses 45 to 55, wherein the sub-field relates to a single die or a functional area of ​​the single die.

[0253] 57. A method according to any of clauses 45 to 56, wherein the first component spatial distribution is used to control one or both of stage positioning control within the lithographic apparatus or projection lens control within the lithographic apparatus.

[0254] 58. A computer program comprising program instructions operable when run on suitable equipment to perform the method according to any of clauses 45 to 57.

[0255] 59. A non-transitory computer program carrier comprising a computer program according to clause 58.

[0256] 60. The method of clause 35, wherein the first component spatial distribution is associated with a first exposure used to pattern the first layer, and the second component spatial distribution is associated with a second exposure used to pattern the first layer.

[0257] 61. The method of clause 60, wherein the first component spatial distribution is associated with a first reticle associated with the first exposure, and the second spatial distribution is associated with a second reticle associated with the second exposure.

[0258] 62. The method of clause 60, wherein the first exposure is associated with a first set of subfields or dies on a reticle and the second exposure is associated with a second set of subfields or dies on the reticle or another reticle.

[0259] 63. The method of clause 61, wherein the first reticle and the second reticle include a first die region associated with a sub-field having no control requirements applied or having a reduced control requirement applied and a second die region having a control requirement applied.

[0260] 64. The method of clause 63, wherein the first die region on the first reticle is associated with a different sub-field than the first die region on the second reticle.

[0261] 65. A method for controlling a photolithography process for exposing a pattern on an exposure field, the method comprising:

[0262] obtaining an initial spatial distribution associated with a spatial variation of a performance parameter across the exposure field;

[0263] The initial spatial distribution is decomposed into at least a first component spatial distribution for controlling the lithography device to correct the spatial variation of the performance parameter across a first plurality of subfields in a first exposure step, and a second component spatial distribution for controlling the lithography device to correct the spatial variation of the performance parameter across a second plurality of subfields in a second exposure step, wherein control of the lithography process using the first component spatial distribution and the second component spatial distribution during their corresponding exposure steps mitigates the spatial variation of the performance parameter across the exposure fields.

[0264] 66. The method of clause 65, wherein the first plurality of subfields and the second plurality of subfields are associated with a die layout on a patterning device such as a reticle or mask.

[0265] 67. A method according to clause 65 or 66, wherein the first plurality of sub-fields are associated with a first patterning device and the second plurality of sub-fields are associated with a second patterning device.

[0266] 68. A method according to clause 65 or 66, wherein the first plurality of subfields and the second plurality of subfields are associated with a die layout on a patterning device, wherein the second plurality of subfields are associated with a different position of the patterning device relative to the substrate compared to a position of the patterning device associated with the first plurality of subfields relative to the substrate used in the lithography process.

[0267] 69. A method according to any of clauses 65 to 68, wherein the controlling of the lithography process is controlling a stage and / or a projection lens actuator of a lithography apparatus used during the lithography process.

[0268] 70. A method according to clause 65 or 66, wherein the first plurality of sub-fields are associated with a first group of tube cores on a patterning device, and the second plurality of sub-fields are associated with a second group of tube cores on the patterning device or another patterning device, wherein the relative positions of the first group of tube cores with respect to the patterning device are staggered with respect to the positions of the second group of tube cores with respect to the patterning device or the other patterning device.

[0269] 71. A lithography tool comprising a control interface, wherein the control interface is configured to receive information associated with a definition of a subset of subfields included in an exposure field, the subset being associated with subfields scheduled to be exposed by the lithography tool in an exposure step.

[0270] Although a patterning device in the form of a physical mask has been described, the term "patterning device" in this application also includes a data product that conveys a pattern in digital form, such as used in conjunction with a programmable patterning device.

[0271] Although specific reference has been made above to the use of embodiments of the invention in the context of optical lithography, it will be appreciated that the invention may be used in other applications, such as imprint lithography, and is not limited to lithography where the context permits. In imprint lithography, the topography in the patterning device defines the pattern created on the substrate. The topography of the patterning device may be pressed into a resist layer provided to the substrate, and the resist is then cured on the resist layer by applying electromagnetic radiation, heat, pressure, or a combination thereof. After the resist has cured, the patterning device is removed from the resist to leave a pattern therein.

[0272] The terms "radiation" and "beam" used in connection with lithographic apparatus encompass all types of electromagnetic radiation, including ultraviolet (UV) radiation (e.g., having a wavelength of or approximately 365, 355, 248, 193, 157 or 126 nm) and extreme ultraviolet (EUV) radiation (e.g., having a wavelength in the range of 5-20 nm), as well as particle beams such as ion beams or electron beams.

[0273] The term "lens," where the context permits, may refer to any one or combination of various types of optical components, including refractive, reflective, magnetic, electromagnetic, and electrostatic optical components.

[0274] The above description of specific embodiments will reveal the general nature of the invention so fully that others can easily modify and / or adapt the various applications of these specific embodiments by applying the knowledge in the art without departing from the general concept of the invention, without undue experimentation. Therefore, based on the teachings and guidance presented herein, such modifications and adaptations are intended to be within the meaning and scope of the equivalents of the disclosed embodiments. It should be understood that the wording or terminology herein is intended to be described by example rather than limiting, so that the terms or wording of this specification will be interpreted by those skilled in the art based on the teachings and guidance.

[0275] The breadth and scope of the present invention should not be limited by any of the above-described exemplary embodiments, but should be defined only in accordance with the following claims and their equivalents.

Claims

1. A method comprising: obtaining an error distribution between a first layer and a second layer across subfields within an exposure field; as well as A first correction distribution for the first layer and a second correction distribution for the second layer are determined by a hardware computer system, wherein the first correction distribution and the second correction distribution are added to mitigate the error distribution across the subfield and are jointly optimized to achieve desired imaging quality across both the first layer and the second layer. The method of claim 1 , wherein the error distribution comprises an overlapping error distribution.

3. The method of claim 1, further comprising adjusting a correction profile of a patterning device used in exposure of the first layer and / or the second layer. The method of claim 1 , wherein the determining evaluates movement characteristics of a substrate to which the first layer and the second layer are applied. The method of claim 1 , wherein the determining comprises optimizing one or more radiation conditions. The method of claim 4 , wherein the optimizing comprises multiple focus level optimizations. 7 . A computer-readable recording medium comprising program instructions stored therein, the instructions being configured to cause a computer system to perform at least the method according to claim 1 when executed by the computer system.

8. A method comprising: obtaining an error distribution extending across a plurality of subfields along a scanning direction; as well as A reticle correction is determined by a hardware computer system across at least two adjacent subfields of the plurality of subfields, wherein the reticle correction is based on an expected improvement in continuity of the error distribution between the adjacent subfields.

9. The method of claim 8, wherein the reticle correction is configured to achieve continuity of the error distribution across the adjacent subfields by substantial inversion and / or overcorrection of the error distribution.

10. A method according to claim 8, wherein a portion of the mask correction causes an error in its corresponding subfield of the adjacent subfield, which error can be more easily corrected by the lithographic equipment from its current state or configuration after correcting a previous subfield of the adjacent subfield, wherein the previous subfield has been subjected to another portion of the mask correction. The method of claim 8 , wherein the error distribution comprises an overlapping error distribution.

12. The method of claim 8, further comprising co-optimizing the reticle correction with a lithography tool control profile.

13. The method of claim 8, further comprising decomposing the reticle correction over multiple layers of a substrate.

14. A computer-readable recording medium comprising program instructions stored therein, the instructions being configured to cause a computer system to perform at least the method according to claim 8 when executed by the computer system.

15. A method for providing an APC control input to an APC controller and a subfield control input to a lithographic apparatus, the method comprising: obtaining a fingerprint of parameters measured across the substrate; Decomposing the fingerprint into a global component and a subfield component; determining a jointly optimized control strategy using the global component and the sub-field components and a control characteristic of the lithographic apparatus; as well as Based on the common optimization control strategy, the APC control input and the sub-field control input are determined.

16. The method of claim 15, wherein the parameter includes overlap and the control strategy is configured to control overlap.

17. The method of claim 15, wherein the APC control and sub-field control inputs are configured to be applied only to a subset of the dies of a substrate.

18. The method of claim 15, wherein the determination of the co-optimization control strategy comprises determining appropriate weightings of the global component data and sub-field component data or corresponding corrections thereof.

19. The method of claim 15, comprising fitting a global model before fitting the residuals based on the subfields.

20. A computer-readable recording medium comprising program instructions stored therein, the instructions being configured to cause a computer system to perform at least the method according to claim 15 when executed by the computer system.

Images