Select a collection of locations associated with measurements or features on a substrate
By selecting the optimal measurement or feature site set on the substrate in the lithography device, using simulated annealing and evolutionary algorithm optimization solutions, the accuracy and focus uniformity of measurement and placement features in the lithography device are solved, and the semiconductor is improved Device quality and output.
Patent Information
- Application Number
- CN202110630682.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2017-03-13
- Filing Date
- 2017-09-21
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2037-09-21
AI Technical Summary
When measuring and placing substrate features, it is difficult for existing lithography equipment to achieve sufficient accuracy and focus uniformity without affecting the production volume, resulting in inaccurate placement of patterns on the substrate, affecting the quality of semiconductor devices and Yield.
By selecting the optimal set of measurement or feature locations on the substrate, using simulated annealing methods and evolutionary algorithms to optimize candidates, combined with constraints such as measurement time, feature uniformity and focus indicators, the optimal measurement or placement position is determined to improve the Accuracy and efficiency of pattern placement.
It is achieved without increasing measurement time and reducing throughput, improving the accuracy and focus uniformity of substrate feature measurement, reducing errors in processing fingerprints, and improving the quality and output of semiconductor devices.
Smart Images

Figure CN113341659B_ABST
Abstract
Description
[0001] This application is a divisional application of the patent application with application number 201780063402.8 and invention name “Selecting a collection of locations associated with measurements or features on a substrate” (international application date is 2017-09-21, international application number is PCT / EP2017 / 073866), whose original applicant is “ASML Netherlands B.V.” and whose entry date into the Chinese national phase is April 12, 2019. Technical Field
[0002] The invention relates to a method of selecting an optimal solution comprising an optimal set of locations associated with measurements or features on a substrate, a related apparatus and a computer program. Background Art
[0003] 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 integrated circuits (ICs). In this case, a pattern forming device, alternatively referred to as a mask or reticle, can be used to produce a circuit pattern to be formed on a single layer of the IC. The pattern can be transferred to a target portion (e.g., including a portion of a die, one or more dies) on a substrate (e.g., a silicon wafer). Typically, the transfer of the pattern is performed by imaging the pattern onto a layer of radiation-sensitive material (resist) disposed on the substrate. Typically, a single substrate will contain a network of adjacent target portions that are patterned in succession. Known lithographic apparatus include so-called steppers and so-called scanners; in a stepper, each target portion is irradiated by exposing the entire pattern to the target portion at once; each target portion is irradiated by scanning the pattern along a given direction (the "scanning" direction) while scanning the substrate in a direction parallel or antiparallel to the direction by a radiation beam. In addition, the pattern can be transferred from the pattern forming device to the substrate by imprinting the pattern onto the substrate.
[0004] Regardless of which type of equipment is used, accurate placement of patterns on a substrate is a major challenge for reducing the size of circuit components and other products that can be produced by lithography. Specifically, the challenge of accurately measuring features on a substrate that has already been mounted is a critical step in being able to position successive layers of superimposed features accurately enough to produce working devices at high yields. Typically, so-called overlay should be achieved within tens of nanometers in today's submicron semiconductor devices, down to within a few nanometers in the most critical layers.
[0005] Another challenge for lithographic equipment is to expose all patterns on the substrate with sufficient focus uniformity. This is important so that the yield of the entire substrate does not suffer from defocus failures at the edge far from the substrate. The printed image on the substrate has sufficient contrast only in a limited (vertical) range about the optimal exposure focus. Out-of-focus exposure may not only lead to reduced contrast, but also may cause changes in the critical dimensions of the pattern in some cases. Defects in semiconductor processing equipment and processes (such as lithography, etching, baking, polishing and annealing) may introduce defects in the entire substrate. Patterns with defects are called process fingerprints, and such defects produce process distortions that may cause overlap or focus errors. The process fingerprint is usually directly characterized based on the correctable parameters of the lithographic equipment.
[0006] Therefore, modern lithography equipment involves extensive measurement or "mapping" operations before the step of actually exposing or otherwise patterning the substrate at the target location. These operations are time-consuming, limiting the throughput of lithography equipment, thereby increasing the unit cost of semiconductor integrated circuits or other products.
[0007] By measuring or placing features at multiple locations on the substrate, a process fingerprint such as a focus or overlap fingerprint is determined.
[0008] As pattern features become smaller and overlay performance requirements become more demanding, so-called advanced alignment and focus models have been and continue to be developed to model and more accurately correct for distortions of the "wafer grid". These advanced models rely on measuring a greater number of target features across the substrate. However, ultimately only a limited number of available target features can be measured without unduly limiting the throughput of the lithography process as a whole. Focus variations across the substrate occur with a high frequency, which requires that a certain number of target features be placed and measured per exposure image in order to capture the focus fingerprint with sufficient accuracy.
[0009] Selecting an optimal set of locations on the substrate to measure or place features can improve the accuracy of determining the process fingerprint.
[0010] A sampling algorithm across the wafer or a sample plan optimizer can be used to select the optimal set of locations on the substrate for taking measurements. The sampling algorithm determines fingerprints such as overlap across the substrate or substrate deformation. Constraints are that these locations need to be distributed with sufficient density across the entire substrate. Open areas should be avoided. The number of sample points should be limited to reduce the impact on measurement time. The sampling algorithm should be fast enough to be practical in a production environment.
[0011] A field sampling algorithm or a mark placement optimizer can be used to select the optimal set of sites on the substrate for placing target features. The field sampling algorithm has other requirements. It needs to facilitate the determination of field specific fingerprints, such as focus. Constraints are the allowed focus mark sites, the maximum number of marks, etc.
[0012] Known methods are unable to achieve across-wafer and across-field sampling while being flexible enough to handle the above constraints. Summary of the invention
[0013] The inventors have discovered a method that can implement both across-wafer and across-field sampling while being flexible enough to handle the above constraints and others.
[0014] According to a first aspect of the present invention, there is provided a method for selecting an optimal solution, the optimal solution comprising an optimal set of locations on a substrate, the locations being associated with positions configured for performing measurement or placement of features, the method comprising the following steps: defining a first candidate solution comprising a first set of locations; defining a second candidate solution comprising a second set of locations based on a modification of coordinates in a solution domain of the first candidate solution, wherein the modification of the coordinates comprises one or more operations selected from repetition, mutation and cross-over; and selecting the first candidate solution and / or the second candidate solution as the optimal solution based on constraints associated with one or more of the following: available locations that can be used to place features on the substrate, available measurement time, a minimum amount of measurements or features associated with an area on the substrate, and a desired degree of uniformity of measurement or feature locations when distributed across the substrate.
[0015] According to a second aspect of the present invention, a method for selecting an optimal solution is provided, the optimal solution comprising an optimal set of locations on a substrate, which are associated with positions configured to perform measurement or placement of features, the method comprising the following steps: defining a first candidate solution comprising a first set of locations; defining a second candidate solution comprising a second set of locations based on a modification of coordinates in a solution domain of the first candidate solution; and selecting the first candidate solution and / or the second candidate solution as the optimal solution based on constraints associated with one or more of the following: available locations that can be used to place features on the substrate, available measurement time, a minimum amount of measurements or features associated with an area on the substrate, and a desired degree of uniformity of measurement or feature locations when distributed across the substrate, wherein the first candidate solution and the second candidate solution are states in a search space comprising the solution domain, and the steps are performed according to a simulated annealing method.
[0016] According to a third aspect of the present invention, there is provided a computer program comprising computer readable instructions, wherein when the computer instructions are executed on a suitable computer device, the computer device is caused to perform the method according to the first aspect.
[0017] According to a fourth aspect of the present invention, there is provided a computer program product comprising the computer program according to the third aspect.
[0018] According to a fifth aspect of the present invention, there is provided an apparatus, which is particularly suitable for performing the steps in the method according to the first aspect. The apparatus may be configured as a lithographic apparatus operable to perform a lithographic process on the substrate. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Embodiments of the present invention will now be described by way of example only with reference to the accompanying schematic drawings. In the drawings, corresponding reference numerals indicate corresponding parts; in the drawings:
[0020] Figure 1 depicts a lithographic apparatus according to an embodiment of the invention;
[0021] Figure 2 Schematically shows Figure 1 multiple stages of measurement and exposure processing in the device;
[0022] Figure 3 is a flow chart illustrating a method according to an embodiment of the present invention;
[0023] Figure 4 is a flow chart illustrating a method according to another embodiment of the present invention;
[0024] FIG. 5A to FIG. 5C depicts allowed marker positions, a conventionally optimized marker layout, and a marker layout optimized according to an embodiment of the present invention;
[0025] Figure 5D and Figure 5E The focus maps before and after correction are depicted respectively.
[0026] Figure 6 is a graph of the sum of standard deviations (sigmas) of residual values optimized according to an embodiment of the present invention;
[0027] Figure 7 is a flow chart illustrating a method according to another embodiment of the present invention;
[0028] Figure 8 Depicts a typical non-uniform sampling triangular sampling scheme for a conventional sampling scheme optimizer;
[0029] Fig. 9depicts a triangular sample solution produced by a method according to an embodiment of the present invention; and
[0030] Fig.10 Computer system hardware suitable for implementing the disclosed methods is shown.
[0031] Fig.11 The concept of using the Pareto front according to an embodiment of the present invention is illustrated. DETAILED DESCRIPTION
[0032] Figure 1 A lithographic apparatus LA according to an embodiment of the invention is schematically depicted.
[0033] The equipment includes:
[0034] i. An illumination system (illuminator) IL configured to condition a radiation beam B (eg UV radiation or EUV radiation).
[0035] ii. a support structure (eg, mask table) MT configured to support a patterning device (eg, mask) MA and connected to a first positioner PM configured to accurately position the patterning device according to certain parameters;
[0036] iii. a substrate table (e.g., wafer stage) WTa or WTb configured to hold a substrate (e.g., a resist-coated wafer) W and connected to a second positioner PW configured to accurately position the substrate according to certain parameters; and
[0037] iv. a projection system (eg, a refractive projection lens system) PS configured to project the pattern imparted to the radiation beam B by the patterning device MA onto a target portion C of the substrate W (eg, comprising one or more dies).
[0038] The illumination system may include various types of optical components, such as refractive, reflective, magnetic, electromagnetic, electrostatic or other types of optical components, or any combination thereof, to direct, shape or control radiation.
[0039] The support structure supports the pattern forming device, that is, bears the weight of the pattern forming device. The mask support structure holds the pattern forming device in a manner that depends on the orientation of the pattern forming device, the design of the lithographic equipment, and other conditions such as whether the pattern forming device is held in a vacuum environment. The support structure can use mechanical, vacuum, electrostatic or other clamping techniques to hold the pattern forming device. The support structure can be a frame or a table, for example, it can be fixed or movable as required. The support structure can ensure that the pattern forming device is located in a desired position, for example, relative to a projection system. It can be considered that any term "reticle" or "mask" used herein can be considered to be synonymous with the more general term "pattern forming device".
[0040] The term "patterning device" as used herein should be broadly interpreted as referring to any device that can be used to impart a radiation beam with a pattern in its cross-section so as to form a pattern in a target portion of the substrate. It should be noted that the pattern imparted to the radiation beam may not exactly correspond to the desired pattern in the target portion of the substrate (for example if the pattern includes phase-shifting features or so called assist features). Typically, the pattern imparted to the radiation beam will correspond to a specific functional layer in a device being formed in the target portion (e.g. an integrated circuit).
[0041] The patterning device may be transmissive or reflective. Examples of patterning devices include masks, programmable mirror arrays, and programmable LCD panels. Masks are well known in photolithography and include mask types such as binary masks, alternating phase shift masks, attenuated phase shift masks, and various hybrid mask types. An example of a programmable mirror array employs a matrix arrangement of small mirrors, each of which can be independently tilted to reflect an incident radiation beam in different directions. The tilted mirrors impart a pattern to the radiation beam reflected by the matrix of mirrors.
[0042] The term "projection system" as used herein should be broadly interpreted as including any type of projection system, including refractive, reflective, catadioptric, magnetic, electromagnetic and electrostatic optical systems or any combination thereof, as appropriate to the exposure radiation used or other factors such as the use of immersion liquid or the use of a vacuum. It is intended that any use of the term "projection lens" herein be considered synonymous with the more general term "projection system".
[0043] As shown here, the lithographic apparatus is of the transmissive type (eg employing a transmissive mask). Alternatively, the lithographic apparatus may be of the reflective type (eg employing a programmable mirror array of the type described above, or employing a reflective mask).
[0044] The lithographic apparatus may be of a type having two substrate tables (dual stage) or more substrate tables (and / or two or more mask tables). In such a "multi-stage" machine, the multiple stages may be used in parallel, or preparatory steps may be performed on one or more stages while one or more other stages are being used for exposure. Although the invention disclosed herein may be used independently, it may also provide additional functionality during the pre-exposure measurement phase of a single-stage or multi-stage apparatus.
[0045] The lithographic apparatus is of a type in which at least a portion of the substrate may be covered by a liquid having a relatively high refractive index, such as water, so as to fill an immersion space between the projection system and the substrate. Immersion liquid may also be applied to other spaces in the lithographic apparatus, such as the space between the mask and the projection system. Immersion techniques are well known in the art and are used to increase the numerical aperture of the projection system. The term "immersion" as used herein does not mean that structures such as the substrate must be immersed in the liquid, but rather "immersion" only means that the liquid is located between the projection system and the substrate during exposure.
[0046] Reference Figure 1 , the illuminator IL receives a radiation beam from a radiation source SO. The radiation source and the lithographic apparatus may be separate entities (for example when the radiation source is an excimer laser). In this case, the radiation source is not considered to form part of the lithographic apparatus, and the radiation beam is transferred from the source SO to the illuminator IL with the aid of a beam delivery system BD comprising, for example, suitable directional mirrors and / or a beam expander. In other cases, the radiation source may be an integral part of the lithographic apparatus (for example when the radiation source is a mercury lamp). The source SO and the illuminator IL, and the beam delivery system BD provided if necessary, may be referred to together as a radiation system.
[0047] The illuminator IL may include an adjuster AD for adjusting the angular intensity distribution of the radiation beam. Typically, at least the outer and / or inner radial extent (generally referred to as -outer and -inner, respectively) of the intensity distribution in a pupil plane of the illuminator may be adjusted. Furthermore, the illuminator IL may include various other components, such as an integrator IN and a condenser CO. The illuminator may be used to adjust the radiation beam so as to have a desired uniformity and intensity distribution in its cross-section.
[0048] The radiation beam B is incident on a patterning device (e.g. mask MA) held on a support structure (e.g. mask table MT) and is patterned by the patterning device. After passing through mask MA, the radiation beam B passes through a projection system PS which focuses the radiation beam onto a target portion C of the substrate W. With the aid of a second positioner PW and a position sensor IF (e.g. an interferometer arrangement, a linear encoder or a capacitive sensor), the substrate table WTa / WTb can be precisely moved, for example in order to position a different target portion C in the path of the radiation beam B. Similarly, the first positioner PM and a further position sensor ( Figure 1 The first positioner PM (not explicitly shown) is used to accurately position the mask MA relative to the path of the radiation beam B. In general, movement of the mask table MT can be achieved with the help of a long-stroke module (coarse positioning) and a short-stroke module (fine positioning) which form part of the first positioner PM. Similarly, movement of the substrate table WTa / WTb can be achieved using a long-stroke module and a short-stroke module which form part of the second positioner PW. In the case of a stepper (as opposed to a scanner), the mask table MT may be connected to the short-stroke actuator only, or may be fixed. Mask MA and substrate W may be aligned using mask alignment marks M1, M2 and substrate alignment marks P1, P2. Although the substrate alignment marks shown in the figures occupy dedicated target portions, they may be located in the spaces between target portions (these are known as scribe alignment marks). Similarly, in the case where more than one die is arranged on the mask MA, the mask alignment marks may be located between the dies.
[0049] The device shown in the figure can be used in at least one of the following modes:
[0050] 1. In step mode, while the mask table MT and substrate table WTa / WTb are held substantially stationary, an entire pattern imparted to the radiation beam is projected at once onto a target portion C (i.e. a single static exposure). The substrate table WTa / WTb is then moved in the X-direction and / or Y-direction so that a different target portion C can be exposed. In step mode, the maximum size of the exposure field limits the size of the target portion C imaged in a single static exposure.
[0051] 2. In scan mode, the mask table MT and the substrate table WTa / WTb are scanned synchronously while a pattern imparted to the radiation beam is projected onto a target portion C (i.e. a single dynamic exposure). The speed and direction of the substrate table WTa / WTb relative to the mask table MT may be determined by the (or less) magnification and image reversal characteristics of the projection system PS. In scan mode, the maximum size of the exposure field limits the width of the target portion (in the non-scanning direction) in a single dynamic exposure, while the length of the scanning motion (and the size of the exposure field) determines the height of the target portion (in the scanning direction).
[0052] 3. In another mode, the mask table MT holding the programmable patterning device is held substantially stationary and the substrate table WTa / WTb is moved or scanned while a pattern imparted to the radiation beam is projected onto a target portion C. In this mode, a pulsed radiation source is typically employed and the programmable patterning device is updated as required after each movement of the substrate table WTa / WTb or between successive radiation pulses during a scan. This mode of operation can be readily applied to maskless lithography using a programmable patterning device, such as a programmable mirror array of the type described above.
[0053] Combinations and / or variations on the above described modes of use or entirely different modes of use may also be employed.
[0054] The lithographic apparatus LA in this example is of a so-called dual-platform type having two substrate tables WTa, WTb and two stations - an exposure station and a measuring station - between which the substrate table can be exchanged. While a substrate on one substrate table is being exposed at the exposure station EXP, another substrate can be loaded onto another substrate table at the measuring station MEA, so that various preparatory steps can be performed. The preparatory steps may include using a horizontal sensor LS to map the surface of the substrate and using an alignment sensor AS to measure the position of alignment marks on the substrate. This enables a substantial increase in the throughput of the apparatus. If the position sensor IF is unable to measure the position of the substrate table when the substrate table is at the measuring station and at the exposure station, a second position sensor may be provided so as to be able to track the position of the substrate table at the two stations. The present invention can be applied to an apparatus having only one substrate table or having more than two substrate tables.
[0055] The device also includes a lithographic equipment control unit LACU, which controls all movements and measurements of the various actuators and sensors described. LACU also includes signal processing and data processing capabilities to implement the desired calculations related to the operation of the device. In practice, the control unit LACU will be implemented as a system with many subunits, each subunit handling the real-time data acquisition, processing and control of the subsystems or components in the device. For example, a processing subsystem can be dedicated to the servo control of the substrate positioner PW. Discrete units can even handle coarse actuators and fine actuators, or different axes. Another unit can be dedicated to the readout of the position sensor IF. The overall control of the equipment can be controlled by a central processing unit, which communicates with these subsystem processing units, with operators and with other equipment involved in the lithographic manufacturing process.
[0056] Figure 2 Shown is the exposure Figure 1 The invention relates to known steps for measuring a target portion (e.g. a die) on a substrate W in a dual-stage apparatus of a photolithography system. The left side within the dashed box are steps performed at the measurement station MEA, while the right side shows steps performed at the exposure station EXP. Sometimes, one of the substrate tables WTa, WTb will be at the exposure station, while the other will be at the measurement station, as described above. For the purpose of this description, it is assumed that the substrate W has already been loaded into the exposure station. At step 200, a new substrate W' is loaded onto the apparatus by a mechanism not shown. The two substrates are processed in parallel in order to increase the throughput of the photolithography apparatus. Let's start with the newly loaded substrate W', which may be a previously unprocessed substrate to which a new resist is applied for the first exposure in the apparatus. However, typically, the photolithography process described will be only one step in a series of exposure and processing steps, so that the substrate W' has passed through the apparatus and / or other photolithography apparatus several times, and may also undergo subsequent processing.
[0057] Previous and / or subsequent processing (as just mentioned) may be performed in other lithographic equipment, and may even be performed in different types of lithographic equipment. For example, some layers in a device manufacturing process that are extremely demanding in parameters such as resolution and overlay may be performed in more advanced lithographic tools than other layers that are less demanding. Thus, some layers may be exposed in an immersion lithography tool, while other layers are exposed in a "dry" tool. Some layers may be exposed in a tool operating at a DUV wavelength, while other layers are exposed using EUV wavelength radiation.
[0058] exist Figure 2In the embodiment of the present invention, at 202, the alignment of the substrate relative to the substrate table WTa / WTb is measured and recorded using alignment measurements using substrate mark P1 and an image sensor (not shown). In addition, a number of alignment marks across the substrate W' are measured to establish a "wafer grid" that very accurately maps the distribution of the marks across the substrate, including any distortion relative to the nominal rectangular grid. At step 204, a substrate height map with respect to XY position is also measured for accurate focusing of the exposed pattern.
[0059] When loading substrate W', receive configuration data 206, this configuration data defines the exposure to be performed, and also defines the properties of the substrate and the previously generated pattern and the pattern to be generated on the wafer. The measurement of the substrate position, substrate grid and height map obtained at 202, 204 is newly added to these configuration data, so that the complete set of configuration and measurement data 208 can be transferred to the exposure platform. The measurement of alignment data includes, for example, the X position and Y position of the alignment target formed in a fixed or nominally fixed relationship with the product pattern of the product processed for lithography. These alignment data obtained just before exposure are combined and interpolated to provide the parameters of the alignment model. These parameters and alignment model will be used during the exposure operation to correct the position of the pattern applied in the current lithography step. Known alignment models may include four, five or six parameters, which together define the translation, rotation and scaling of the "ideal" grid at different sizes. As further described below, advanced models using more parameters are known.
[0060] At 210, substrates W' and W are exchanged so that the measured substrate W' becomes substrate W and enters the exposure station EXP. This exchange is performed by exchanging supports WTa and WTb within the device so that substrates W, W' remain accurately clamped and positioned on those supports to maintain the relative alignment between the substrate stage and the substrate itself. Therefore, once these stages have been exchanged, in order to use the measurement information 202, 204 for substrate W (i.e., the previous W') to control the exposure step, it is necessary to determine the relative position between the projection system PS and the substrate stage WTb (i.e., the previous WTa). At step 212, mask alignment is performed using mask alignment marks M1, M2. In steps 214, 216, 218, scanning movements and radiation pulses are applied to consecutive target locations across substrate W in order to complete the exposure of several patterns. By using the alignment data and height map obtained at the measurement station in performing the exposure step, these patterns are accurately aligned relative to the desired locations, more specifically, accurately aligned relative to the features previously placed on the same substrate. At step 220, the exposed substrate, now labeled W", is removed from the apparatus to undergo etching or other processing according to the exposed pattern.
[0061] Selecting the optimal set of locations for measurement or characterization is an optimization problem.
[0062] Metaheuristics are high-level procedures or heuristics designed to find, generate, or select heuristics (partial search algorithms) that provide sufficiently good solutions to optimization problems, especially in the presence of incomplete or imperfect information or limited computing power.
[0063] Evolutionary algorithms (EAs) are general population-based meta-heuristic optimization algorithms.
[0064] A genetic algorithm (GA) is a type of EA. In a GA, a population of candidate solutions to an optimization problem (called individuals, creations, or phenotypes) evolve towards better solutions. Each candidate solution has a set of properties (its chromosome or genotype) represented by coordinates in a solution domain, which can mutate and change. These solutions can be represented in binary as strings of 0s and 1s, but other encodings are possible.
[0065] Simulated annealing (SA) is a probabilistic technique for approximating the global optimization of a given function. SA is a metaheuristic method for approximating the global optimization in a large search space or solution domain.
[0066] Tabu search is another meta-heuristic search method that uses local search methods for mathematical optimization.
[0067] Genetic algorithms maintain a pool of solutions, rather than just one. New candidate solutions are generated not only by "mutation" (as in SA), but also by "recombination" of two solutions from the reservoir. Similar to the criteria used in SA, probabilistic criteria are used to select candidates for duplication, mutation or combination (by exchange), and for discarding excess solutions from the reservoir.
[0068] In mathematical optimization, a cost function or loss function is a function that maps the value of an event or one or more variables to a real number that intuitively represents some "cost" associated with the event. Optimization problems seek to minimize the cost function. The objective function can be the cost function, or its negative value (sometimes called a merit function, reward function, profit function, utility function, fitness function, etc.) - in which case it is to be maximized.
[0069] Constraints are conditions that the solution to an optimization problem must satisfy.
[0070] Figure 3 is a flow chart illustrating a method for selecting an optimal solution including an optimal set of locations associated with measurements or features on a substrate according to an embodiment of the present invention.
[0071] The method has the following steps:
[0072] 302: Define constraints and optionally one or more cost functions. These constraints and cost functions are discussed in more detail below. The cost function can calculate the difference between the modeled fingerprint and the measured data. The cost function is used to assign a value to an individual (candidate solution) for comparison with other individuals (candidate solutions).
[0073] 304: Define a search space. For example, the search space may be locations in the field where focus marks can be placed. The search space may be a feasible set of candidate solutions that satisfy all constraints. It may be a complete set of measurement locations (i.e., all feasible locations). The boundaries of the search space may be defined by the size of the wafer.
[0074] 306: Define a first candidate solution including a first set of locations. The first candidate solution is a set of one or more locations, such as marking or measurement locations in a field or on a wafer. Each candidate solution can be defined by a coordinate in a search space. The first candidate solution can be an individual in a population of candidate solutions. In this case, the population is a first generation.
[0075] This first candidate solution may be based on the generation of random or pseudo-random coordinates in the search space.
[0076] The first candidate may be based on knowledge of constraints. Such knowledge may include the layout and layers used to make a product (e.g., a particular integrated circuit). The knowledge may include configurations and processes used to make the product that constrain where the marks or measurement points should be placed. For example, the constraints may be one or more of the following: no more than 5 marks per row; one mark per die; 10 marks per field; and a limit on the actual area used by the marks.
[0077] 308: Based on the modification of the coordinates in the search space of the first candidate, a second candidate including a second set of parts in the search space is defined. The modification of the coordinates may include one or more operations of repetition, mutation, and exchange. The second candidate may be an individual in a population of candidates. In this case, the population is the second generation relative to the first candidate as the first generation. Because there are many generations, the first and second may refer to the current generation and the next generation, or the previous generation and the current generation.
[0078] 310: Optionally, determine a value of a cost function associated with a desired measurement accuracy or feature placement on the substrate.
[0079] 312: Select the first candidate solution and / or the second candidate solution as the optimal solution according to the constraints associated with the substrate. Optionally, the selection of the first candidate solution and / or the second candidate solution is further based on the (estimated) value of the cost function.
[0080] Constraints can be associated with the available locations for placing features. For example, marks may be allowed only in scribe lanes or at the edge of the wafer. Marks are constrained to be located on the wafer.
[0081] The constraint condition may be associated with a minimum distance required between a location for placing a feature and a keep-away area of the substrate. Figure 5A Discuss this.
[0082] The constraints may be associated with a minimum amount of measurements or features associated with an area on the substrate.
[0083] Constraints may be associated with a desired degree of uniformity of the measurements or features as they are distributed across the substrate.
[0084] The cost function may be associated with one or more of the following.
[0085] A cost function may be associated with an optimal determination of a focus metric across the substrate. For example, the cost function may calculate the difference between a modeled fingerprint and height data measured across the wafer.
[0086] A cost function may be associated with an optimal determination of substrate distortion characteristics. For example, the cost function may use a level sensor to calculate the difference between a modeled fingerprint and height data measured across the wafer. The model may include a mesh model of the substrate surface.
[0087] A cost function may be associated with the optimal positioning of focus marks across the substrate. For example, the cost function may calculate the difference between a modeled fingerprint and height data measured across the wafer. The model may include a mesh model of the substrate or a function such as a Taylor series.
[0088] The cost function may be associated with the optimal positioning of the alignment mark across the substrate. For example, the cost function may calculate the difference between the modeled fingerprint and the x and y coordinate data measured across the wafer. The model may include a set of polynomial functions with respect to x and y. As an alternative to x and y, radial coordinates and functions may be used.
[0089] A cost function may be associated with the optimal positioning of overlapping marks across the substrate. For example, the cost function may calculate the difference between the modeled fingerprint and the x and y coordinate data measured across the wafer. The model may include a set of polynomial functions with respect to x and y. As an alternative to x and y, radial coordinates and functions may be used.
[0090] A cost function may be associated with the optimal positioning of dose marks across the substrate. For example, the cost function may calculate the difference between a modeled fingerprint and line thickness data measured across the wafer. The model may include a set of polynomial functions related to line thickness.
[0091] The cost function may be associated with an optimal number of measurements or features taking into account the measurement time. For example, the cost function may include a set of polynomial functions as a function of the measurement time.
[0092] The cost function may be associated with a quality parameter of a fitting process performed on a pattern of observables defined at one or more of the locations associated with the measurements and / or feature locations.
[0093] A cost function may be associated with an area on the substrate.
[0094] 314: If the constraints are satisfied and (optionally) the value of the cost function has converged, or if a number of iterations has been reached, the iteration ends; otherwise, control passes back to step 308 for another iteration.
[0095] The steps of the method may be performed according to an evolutionary approach, in which case the first candidate and the second candidate are individuals in a continuous population of candidates.
[0096] The steps in the method may be performed according to a simulated annealing method, in which case the candidate solutions are states in a search space comprising a solution domain.
[0097] Refer to below Figures 4 to 6 In another embodiment described, an evolutionary algorithm for solution optimization (EASO) is a search method for finding field mark locations to be used in the mark placement optimizer and the overlap optimizer (and the dose / CD optimizer). Each of these optimizers facilitates improved judgment of the process fingerprint by selecting one or more optimal solutions.
[0098] The optimal solution provides measurement locations or feature placement locations that improve the accuracy of the determined process fingerprint, such as focus, overlap or dose / CD fingerprint. It can also minimize the number of marks per field and maximize the focus correction per field (i.e., minimize the domain-dependent uncorrectable residuals per field). The domain here can be focus, overlap, or dose / CD. Uncorrectable residuals are residuals that cannot be corrected by setting scanner parameters.
[0099] The algorithm is structured as an input set of data and an evolutionary algorithm (EA) optimization function.
[0100] Figure 4is a flow chart showing a method according to another embodiment of the present invention. The method has the following steps:
[0101] 402: Define the evaluation function and parameters used for mutation and exchange.
[0102] The mutation probability parameter determines how often a chromosome mutation occurs. A mutation probability of 100% causes a change in the entire chromosome. A mutation probability of 0% causes no change in the chromosome.
[0103] The crossover probability parameter determines how often crossovers occur. If there are no crossovers, the offspring is an exact copy of its parent. If there are crossovers, the offspring is made from some parts of the chromosomes of the parent. A crossover probability of 100% results in all offspring being made by crossovers. A crossover probability of 0% results in a complete new generation with an exact copy of the chromosomes from the previous generation, except for mutations.
[0104] 404: Based on a priori design knowledge of the die layout and a defocus fingerprint with an expected behavior that needs to be measured as accurately as possible, the input data are locations in the field where focus marks can be placed. Figure 5A A die layout with allowed marking positions enclosing four spaces left for product is shown. The minimum possible defocus for a given layout with the maximum possible (grid) number of markings is determined. The defocus is then optimized to determine the closest to this value with the minimum number of markings or a fixed number of markings.
[0105] 406: EA sets the group of individuals.
[0106] 408: EA performs evolution based on repetition, mutation and exchange. EA is a broad family of algorithms (belonging to machine learning) that optimize multi-dimensional complex problems. Figure 4 The described method may be improved adding ad-hoc heuristics (i.e., restrictions on placement of markers, use of measurement data, etc.) In alternative embodiments, simulated annealing methods or taboo search may be used with movement to nearest neighbors or random search in the search space.
[0107] 410: Based on an individual, the evaluation function returns the score of the individual. The individual represents a mark position in the field, and its score depends on the number of marks and the maximum correction that can be achieved with these marks. The evaluation function can be based on static or dynamic measurements. Static measurements refer to a specific set of wafers that all belong to the same exposure batch. In this case, the evaluation function returns only the residual value of the correction on the wafers in this batch. The correction is calculated for the measured wafers. Dynamic measurements refer to a set of wafers belonging to a series of batches. In this case, the evaluation function returns the residual value due to the function taking into account different batches. If, for example, there are N batches (from 1 to N), the measurements are from batch 1; although only the correction for these measurements is calculated, the correction applies to all batches: from 1 to N. The evaluation function is then based on the residual values of N batches.
[0108] 412: The EA continues to perform evolution through selection.
[0109] 414: If the value of the score is acceptable, or if a certain number of evolutions has been reached, or if any other relevant stop condition is reached, the iteration ends; otherwise, control passes back to step 408 for another iteration.
[0110] The output of the algorithm is the optimal individual in the population after a finite number of evolutions, i.e., the individual with the minimum number of markers placed in a way that maximizes the accuracy of the focus fingerprint determination. In other words, the algorithm reduces the number of markers to be read and minimizes the wafer distortion after applying wafer correction.
[0111] FIG. 5A to FIG. 5C Graphs depicting allowed mark positions, a conventionally optimized mark layout, and a mark layout optimized according to an embodiment of the present invention, respectively.
[0112] Figure 5A The allowed mark locations in the field are depicted (in a 10x10 layout, the empty space is where 4 dies are placed). The constraints are shown with a minimum distance d between a region R (such as the edge of the wafer) in which no marks are allowed and a mark.
[0113] Figure 5B Depicted is the token layout used by a conventional optimizer (with 16 tokens).
[0114] Figure 5C A marking layout found by a method according to an embodiment of the invention is depicted.
[0115] In an embodiment, the optimized focus metric is 3*standard deviation (3*std) across the wafer, which is referred to as focus uniformity. The goal is to have a minimum value of 3*std after the focus optimizer has been applied to the raw data.
[0116] Figure 5C The optimal layout is Figure 5A A subsample of the full layout is shown. The 3*std of the optimized layout is sensitive to the position of the markers. Fewer markers yield a less optimal estimation of the true field fingerprint.
[0117] In order to fairly compare the reduced (optimized) layouts, the correction potential of the focused model sub-configuration of each layout is compared to the full layout.
[0118] The focus model sub-configuration is the inverse of the modeled fingerprint. Therefore, a less accurate fingerprint will produce a poorly corrected sub-configuration. This causes a higher z-variation of the wafer in the focus model (higher 3*std) after correction.
[0119] For overlay, it is slightly different because there are many overlay marks and the overlay optimizer must choose which overlay marks need to be read to get the best fingerprint estimate. However, this only happens at the wafer level and does not scale down to a single field. So while both (focus and overlay) optimizers look for the minimum 3*std across the wafer, overlay is only about wafer level variations, while focus optimization also needs to consider variations across a single field.
[0120] Figure 5D and Figure 5E The focus maps before and after correction are depicted, respectively. The exposure field is shown as a rectangle. The printed image on the substrate has sufficient contrast only in a limited (vertical) range around the optimal exposure focus, shown as white areas within the circular wafer edge. The black features on the circular wafer represent out-of-focus areas. Figure 5D In the image, the black horizontal stripes are examples of intra-field focus variation. The circular center pattern near the center of the wafer and the edge roll-off around the rim are examples of inter-field focus variation. Together, intra-field and inter-field variations form the focus fingerprint. Figure 5E A corrected focus map with a much larger white area in correct focus is depicted.
[0121] Figure 6 is a graph of the sum of the standard deviations (sigmas) of the residuals optimized according to this embodiment of the invention. The residuals are defined as the original shape minus the correction.
[0122] Figure 6 The sum of the standard deviations (sigmas) of the residual values of one of the initial individuals 604 and the complete population 602 within 100 evolutions i of the method according to an embodiment of the present invention is shown.
[0123] The found marker layout is a good improvement over the conventional optimizer since it halves the measurement time without compromising the correction potential.
[0124] Reference Figure 4 The described method has the following characteristics:
[0125] - It takes into account the constraints present in the marker layout optimizer in order to achieve meaningful focus correction;
[0126] -It uses multiple sets of polynomials required for focusing to describe the field shape;
[0127] - it can minimize the number of markers per field; and
[0128] -It adopts multi-objective optimization evolutionary algorithm.
[0129] Reference Figure 4 The advantages of the described approach are:
[0130] - Increase the “real area” available on the reticle for product features (used to manufacture integrated circuit products);
[0131] - Reduced residual values: exposure will be made at more accurately focused (and overlapped) values; and
[0132] - Reduce the number of markings per field, thus leaving more space for product features and / or other markings (eg overlay / dose / CD / etc).
[0133] Reference Figure 4 The described method advantageously provides an algorithm for determining a minimum number of markers and marker placement within a field that requires minimal a priori user knowledge of product features and can be used for both overlay and focus.
[0134] By combining the sample solution optimizer functionality and the tag layout optimizer functionality, it is possible to implement a reference to the focus Figure 4 Describe the method.
[0135] Refer to below Figures 7 to 9 In another embodiment described, a genetic multi-objective binary optimization is used for the sample solution optimizer.
[0136] Figure 7 is a flow chart showing a method according to this embodiment of the present invention. The method has the following steps:
[0137] 702: Define the cost function and select the genetic algorithm (GA) parameters.
[0138] 704: Define the search space (gene reservoir). Assuming that N positions are "available" as candidate positions, identify the complete candidate set of measurement sites (i.e., all feasible sites). Then, initialize a set of vectors, each of which has a length of N and is completely filled with 0. The ordering within the vectors corresponds exactly to the ordering of the complete design matrix of the experiment (i.e., the design matrix of the complete data set). These vectors are chromosomes. They each "encode" a candidate solution for the optimization problem in the form of a Boolean mask, and the vector consists of 0s and 1s. Wherever 1 is, the corresponding column of the design matrix is included. Wherever 0 is, the corresponding column of the design matrix is ignored (i.e., eliminated from the system of model equations).
[0139] 706: Generate initial population. Assuming there are n locations available for measurement (driven by throughput effects), exactly n zeros out of N zeros are randomly set equal to 1 for each vector. Inclusion or exclusion of a specific measurement location is encoded by 1 or 0, respectively.
[0140] 708: For the GA, decode the "chromosomes" and find the cost of each chromosome. For each of the initialized vectors, estimate the objective function. For this example, the objective function is a combination of D-optimality (minimization of the variance of the estimated parameters) for the selected measurement locations that is very close to the term favorable uniformity combination and penalty. To define uniformity, Delaunay triangulation is used.
[0141] 710: In the "select pairs" step, once the objective function is estimated for each vector, good vectors are kept and bad vectors are discarded. Good and bad can be defined as either side of a threshold of the estimated mesh size. These quantities are configurable, as is the total population size (i.e., the total number of vectors).
[0142] 712: In the "pairing and mutation" step, the discarded vectors are replaced by combining the best vectors into new vectors, similar in nature to what happens via breeding / pairing. The best solutions are arbitrarily paired in pairs of two, and will produce offspring whose vectors are combinations of the vectors of the parents. In addition, to introduce variation, the mutation vector will act on the newly generated offspring, while turning some offspring into 0 or vice versa. Finally, to achieve that there are exactly n 1s (and Nn 0s) in the vector, the if operator checks whether there is a shortage or surplus of 1s in the vector, and therefore fixes this situation (again in a random manner) by placing an appropriate number of 1s to 0s or vice versa.
[0143] 714: In the convergence check step, the stopping criterion typically consists of a maximum number of iterations and a change in the objective function that becomes too small within a certain number of iterations (e.g., the past 100 iterations); otherwise, control is passed back to step 708 for another iteration.
[0144] For the new population, the objective function is again estimated at step 708, and the optimal vector will again produce offspring in the manner described above, and so on.
[0145] The mutation rate starts at a high value (many mutations) and will decrease towards a small number (few mutations) throughout the optimization. The rationale is that in the early stages of optimization, we want a large diversity in order to cover as much of the "search space" as possible (i.e., the space of all combinations of possible solutions to the optimization problem). However, if mutations remain high, the algorithm will never converge towards a solution. The minimum value of the mutation rate can be set to a fixed value, but the value can also be adjusted dynamically to increase / decrease by observing the amount of change in the population (i.e., the standard deviation of its function value relative to the objective function).
[0146] Throughout the optimization process (ie, each iteration), the population size is fixed, as is the amount of 1s and 0s within any vector. The vectors can be viewed as genomes and the algorithm can be viewed as a natural selection method (and the search space is the gene reservoir).
[0147] Figure 8 A typical non-uniform sampling triangular sampling scheme for a conventional sampling scheme optimizer is depicted. Figure 8 The scheme depicted in is only optimized for minimizing model uncertainty. The disadvantage is that when the model it is optimized for is not a good model for describing the actual interference fingerprint, the fingerprint will not be detected. In addition, Figure 8 A large number of fields without any measurement is undesirable in production, since defectivity inspection cannot be performed for those fields.
[0148] Fig. 9 Depicted by reference Figure 7 The described method produces a triangular sample solution. Fig. 9 The scheme is optimized to minimize model uncertainty while constraints are associated with desired degrees of uniformity of measurements or features distributed across the substrate.
[0149] Reference Figure 7 and Fig. 9 The described method is fast and allows for multi-objective optimization. In addition to the optimization of the design of experiments, uniform wafer coverage is also explicitly considered, thus making the optimized sampling plan layout suitable for monitoring and making the optimized sample plan robust to model mismatch.
[0150] Fig.11 An example of using a Pareto front representation of a multi-objective problem is shown. The Pareto front is defined as the relationship between objectives (eg, parameters to be optimized) associated with a multi-objective optimization problem. Fig.11 The Pareto front 1100 of exemplifies the relationship between a first objective of maximizing the focus correction potential on the y-axis (the minimum achievable residual of the focus error after correction) and a second objective of using a minimum number of focus marks (or focus measurements) on the x-axis. The Pareto front can be established by computing a plurality of configurations (e.g., candidate solutions) and determining a measure of the objective for each configuration (e.g., a focus correction residual and a number of marks used to achieve the measure). Fig.11 , each point 1120 represents a certain configuration (distribution) of focus marks (measurements) across the substrate and the residual value after correction when that configuration is applied. It is clear from the figure that some configurations are better suited to determining focus corrections using a certain number of focus marks than others. This is further illustrated by the selection 1110 of a set of focus mark configurations given a constant number / number of marks. The Pareto front 1100 can be approximated by fitting a curve at the lower end of the distribution 1110 (e.g., the lowest residual focus error in terms of number of marks).
[0151] In an embodiment, solving the multi-objective problem using any of the previously described methods (evolutionary algorithms, genetic algorithms, evolutionary algorithms for solution optimization, simulated annealing, tabu search) includes determining a Pareto front, wherein the Pareto front establishes a relationship between the first objective and the second objective. Fig.11 In the example of , the first objective is the focus error residual after focus correction, and the second objective is to minimize the number of focus marks utilized and / or the number of focus measurements performed on the focus marks. However, the Pareto front may be determined to establish any suitable relationship between the first objective and the second objective, for example but not limited to, the first objective may be associated with any one of the overlay error residual, the dose error residual, and the critical dimension error residual, and the second objective may be associated with any one of the number of marks used and the number of measurements performed.
[0152] Fig.11 The one-dimensional Pareto front that establishes the relationship between two objectives is shown. However, in general, the concept of the Pareto front is not limited to a one-dimensional expression, and the Pareto front may indicate the relationship between three or more objectives associated with a multi-objective problem. In this case, the Pareto front may be a surface in two or more dimensions.
[0153] The steps in the method described above can be Figure 1The lithographic apparatus shown in FIG. 1 is automated within a control unit LACU. The unit LACU may include computer components as shown in FIG.
[0154] The computer assembly may be a dedicated computer in the form of a control unit in an embodiment of the component according to the invention, or alternatively a central computer controlling the lithographic projection apparatus. The computer assembly may be configured to load a computer program product comprising computer executable program code. This may enable the computer assembly to control the aforementioned use of the lithographic apparatus by means of embodiments of the level sensor LS and the alignment sensor AS when the computer program product is downloaded.
[0155] The memory 1029 connected to the processor 1027 may include several memory components, such as a hard disk 1061, a read-only memory (ROM) 1062, an electrically erasable programmable read-only memory (EEPROM) 1063, and a random access memory (RAM) 1064. It is not necessary for all of the aforementioned memory components to be present. In addition, the aforementioned memory components do not have to be physically adjacent to the processor 1027 or adjacent to each other. They can be located at a distance.
[0156] The processor 1027 may also be connected to some kind of user interface, such as a keyboard 1065 or a mouse 1066. A touch screen, trackball, voice converter or other interface known to those skilled in the art may also be used.
[0157] The processor 1027 may be connected to a reading unit 1067, which is configured to read data, for example in the form of computer executable code, from a data carrier (e.g., a solid-state drive 1068 or a CDROM 1069), and in some cases store the data on the data carrier. DVDs or other data carriers known to those skilled in the art may also be used.
[0158] The processor 1027 may also be connected to a printer 1070 for printing output data on paper, and to a display 1071 , such as a monitor or a liquid crystal display (LCD), for example, any other type of display known to those skilled in the art.
[0159] The processor 1027 can be connected to a communication network 1072, such as a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), etc., through a transmitter / receiver 1073 responsible for input / output (I / O). The processor 1027 can be configured to communicate with other communication systems via the communication network 1072. In an embodiment of the present invention, an external computer (not shown) (e.g., an operator's personal computer) can log in to the processor 1027 via the communication network 1072.
[0160] Processor 1027 can be implemented as a stand-alone system or as several processing units operating in parallel, wherein each processing unit is configured to perform subtasks of a larger program. The processing unit can also be divided into one or more main processing units and several sub-processing units. Some processing units of processor 1027 can even be located at a distance from other processing units and communicate via communication network 1072. The connection between modules can be wired or wireless.
[0161] The computer system may be any signal processing system with analog and / or digital and / or software technology configured to perform the functions discussed herein.
[0162] Other embodiments of the present invention are disclosed in the following list of embodiments:
[0163] 1. A method for selecting an optimal solution, the optimal solution comprising an optimal set of locations associated with measurements or features on a substrate, the method comprising the following steps:
[0164] defining a first candidate solution including a first set of parts;
[0165] defining a second candidate solution including a second set of parts based on the modification of the coordinates in the solution domain of the first candidate solution; and
[0166] The first candidate solution and / or the second candidate solution are selected as the optimal solution according to the constraint conditions associated with the substrate.
[0167] 2. The method according to embodiment 1, wherein the step of limiting the first candidate scheme is based on the generation of random or pseudo-random numbers.
[0168] 3. The method according to embodiment 1 or embodiment 2, wherein the step of limiting the first candidate solution is based on the recognition of the constraint condition.
[0169] 4. A method according to any preceding embodiment, wherein the modification of the coordinates comprises an operation selected from the following list of operations: duplication; mutation; and exchange.
[0170] 5. The method of any preceding embodiment, wherein the first candidate solution and the second candidate solution are individuals in a continuous population of candidate solutions, and the coordinate step is performed according to an evolutionary method.
[0171] 6. The method of any preceding embodiment, wherein the candidate solutions are states in a search space comprising a solution domain, and the steps are performed according to a simulated annealing method.
[0172] 7. The method of any preceding embodiment, wherein the step of selecting the first candidate and / or the second candidate is further based on a value of a cost function associated with a desired measurement accuracy or feature layout on the substrate.
[0173] 8. The method of embodiment 7, wherein the cost function is associated with an optimal determination of a focus metric across the substrate.
[0174] 9. The method of embodiment 7, wherein the cost function is associated with an optimal determination of substrate distortion characteristics.
[0175] 10. The method of embodiment 7, wherein the cost function is associated with optimal positioning of focus marks across the substrate.
[0176] 11. The method of embodiment 7, wherein the cost function is associated with optimal positioning of alignment marks across the substrate.
[0177] 12. The method of embodiment 7, wherein the cost function is associated with a quality parameter of a fitting process performed on a graph of observables, the observables being confined to one or more of the locations associated with the measurements and / or feature locations.
[0178] 13. The method of any preceding embodiment, wherein the constraints are associated with available locations for placing features.
[0179] 14. The method of any preceding embodiment, wherein the constraint is associated with a minimum distance required between a location for placing a feature and a keep-away zone of the substrate.
[0180] 15. The method of any preceding embodiment, wherein the constraint is associated with a minimum amount of measurements or features associated with an area on the substrate.
[0181] 16. The method of any preceding embodiment, wherein the constraints are associated with a desired degree of uniformity of the measurements or features as they are distributed across the substrate.
[0182] 17. The method of any preceding embodiment, wherein the substrate is a patterning device.
[0183] 18. The method of any preceding embodiment, wherein the substrate is a wafer.
[0184] 19. The method of embodiment 7, wherein the cost function is associated with optimal positioning of overlay marks across the substrate.
[0185] 20. The method of embodiment 7, wherein the cost function is associated with optimal positioning of dose marks across the substrate.
[0186] 21. The method of embodiment 7, wherein the cost function is associated with an optimal number of measurements or features taking into account measurement time.
[0187] 22. The method of any one of embodiments 7 to 21, wherein the cost function is associated with an area on the substrate.
[0188] 23. A method according to any preceding embodiment, wherein a Pareto front is determined, wherein the Pareto front establishes a relationship between a first objective of selecting an optimal solution that includes an optimal set of locations and a second objective of selecting an optimal solution that includes an optimal set of locations.
[0189] 24. The method of embodiment 23, wherein the first target is associated with an optimal determination of a focus index across the substrate, and wherein the first target is associated with an optimal number of measurements or features.
[0190] 25. A computer program comprising computer-readable instructions, wherein when the computer instructions are executed on a suitable computer device, the computer device performs the method as described in embodiments 1 to 24.
[0191] 26. A computer program product comprising the computer program according to embodiment 25.
[0192] 27. An apparatus specially adapted to carry out the steps of the method of embodiments 1 to 24.
[0193] 28. The device according to embodiment 27 can be configured as a lithography device, which can be operated to perform a lithography process on the substrate.
[0194] The term "substrate" as used herein may refer to a wafer or a patterning device, such as a reticle. In photolithography, a reticle has a pattern that is imaged onto a target substrate, such as a wafer.
[0195] Although specific reference is made herein to the use of lithographic apparatus in the manufacture of ICs, it should be understood that the lithographic apparatus described herein may have other applications, such as the manufacture of integrated optical systems, guidance and detection patterns for magnetic domain memories, flat panel displays, liquid crystal displays (LCDs), thin film magnetic heads, etc. Those skilled in the art will recognize that in these alternative application scenarios, any term "wafer" or "field" / "die" used herein may be considered synonymous with the more general term "substrate" or "target portion", respectively. The substrate referred to herein may be processed before or after exposure, such as in a track (a tool that typically applies a layer of resist to a substrate and develops the exposed resist), a metrology tool, and / or an inspection tool. Where applicable, the disclosed content may be applied to such and other substrate processing tools. In addition, the substrate may be processed more than once, such as to produce a multi-layer IC, so that the term "substrate" used herein may also refer to a substrate that already contains multiple processed layers.
[0196] 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 types of optical lithography, such as imprint lithography, and is not limited to optical lithography where circumstances permit. In imprint lithography, the topography in the patterning device defines the pattern produced on the substrate. The topography of the patterning device may be imprinted into a layer of resist provided to the substrate, the resist being cured by applying electromagnetic radiation, heat, pressure, or a combination thereof. The patterning device is removed from the resist, leaving the pattern behind after the resist has cured.
[0197] As used herein, the terms "radiation" and "beam" include all types of electromagnetic radiation, including ultraviolet (UV) radiation (e.g., having a wavelength equal to or approximately 365 nm, 248 nm, 193 nm, 157 nm or 126 nm) and extreme ultraviolet (EUV) radiation (e.g., having a wavelength in the range of 5 nm to 20 nm), as well as particle beams such as ion beams or electron beams.
[0198] 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.
[0199] Although specific embodiments of the present invention have been described above, it should be understood that the present invention may be practiced in other ways than those described. For example, the present invention may be in the form of a computer program comprising one or more sequences of machine-readable instructions describing the methods disclosed above, or a data storage medium (e.g., a semiconductor memory, a magnetic disk, or an optical disk) having the computer program stored therein.
[0200] The above description is intended to be illustrative, not restrictive. Therefore, it will be appreciated by those skilled in the art that modifications may be made to the described invention without departing from the spirit and scope of the claims set forth below. In addition, it should be understood that the structural features or method steps shown or described in any one of the embodiments herein may also be used in other embodiments.
Claims
1. A method for selecting an optimal solution, the optimal solution comprising an optimal set of locations on a substrate, the locations being associated with positions configured for performing measurement or placement of features, the method The following steps are involved: defining a first group of one or more candidates, each candidate of the first group comprising a set of locations; defining a second group including one or more candidates, each candidate of the second group including a set of sites generated by modification of one or more candidates from the first group, wherein the modification includes one or more operations selected from duplication, mutation, and exchange; and selecting one or more candidate solutions from the first group and / or the second group as the optimal solution, Wherein, in the steps of defining the second group and / or selecting the optimal solution, if the candidate solution and / or the optimal solution does not satisfy a constraint associated with a required degree of uniformity of the measurement or feature location when distributed across the substrate, the candidate solution and / or the optimal solution are rejected.
2. The method according to claim 1, in, The step of generating the second group of part sets and / or the step of selecting the optimal solution are performed based on an objective function that is any one of a cost function, an evaluation function, a reward function, a profit function, a utility function, or a fitness function.
3. The method according to claim 1 or 2, in, In the steps of defining the second group and / or selecting the optimal solution, the candidate solutions and / or the optimal solution are rejected if they do not satisfy one or more additional constraints associated with one or more of the following items: available locations for placing features on the substrate, available measurement time, and a minimum number of measurements or features associated with an area on the substrate.
4. The method according to claim 1, in, The step of defining said first set is based on the generation of random or pseudo-random numbers.
5. The method according to claim 1, in, The step of defining said first set is based on the recognition of said constraints.
6. The method according to claim 1, in, The candidate solutions in the first group and / or the second group comprise individuals in a continuous population of candidate solutions, and the steps of the method are performed according to an evolutionary approach.
7. The method according to claim 1, in, The step of selecting the first and / or second group may also depend on the value of a cost function or fitness function associated with a desired measurement accuracy or feature layout on the substrate.
8. The method according to claim 7, in, The cost function or fitness function is associated with an optimal determination of a focus metric across the substrate.
9. The method according to claim 7, in, The cost function or fitness function is associated with an optimal determination of the substrate distortion characteristics.
10. The method according to claim 7, in, The cost function or fitness function is associated with optimal positioning of focus marks, alignment marks, dose marks or overlay marks across the substrate.
11. The method according to claim 7, in, The cost function or goodness of fit function is associated with a quality parameter of a fitting process performed on an observable pattern defined at one or more of the locations associated with the measurements and / or feature locations.
12. The method according to claim 1, in, The substrate is a patterning device or a wafer.
13. The method according to claim 7, in, The cost function or fitness function is associated with the optimal number of measurements or features taking into account the measurement time.
14. The method according to claim 1, further comprising: include: A Pareto front is determined that establishes a relationship between a first objective of selecting an optimal solution that includes an optimal set of locations and a second objective of selecting an optimal solution that includes an optimal set of locations.
15. A computer program comprising computer readable instructions, in, The computer readable instructions, when executed on a suitable computer device, cause the computer device to perform the method of claim 1 .
16. A lithographic apparatus specially adapted to perform the steps of the method as claimed in claim 1.
